Begon et al 2006 - Ecology, from individuals to ecosystems, Notas de estudo de Oceanografia

Baixe Begon et al 2006 - Ecology, from individuals to ecosystems e outras Notas de estudo em PDF para Oceanografia, somente na Docsity! ps [eta apito] as a 7 É Ecology From Individuals to Ecosystems Michael Begon, Colin R. Townsend, and John L. Harper •• ECOLOGY From Individuals to Ecosystems EIPA01 10/24/05 1:36 PM Page i •••• © 1986, 1990, 1996, 2006 by Blackwell Publishing Ltd BLACKWELL PUBLISHING 350 Main Street, Malden, MA 02148-5020, USA 9600 Garsington Road, Oxford OX4 2DQ, UK 550 Swanston Street, Carlton, Victoria 3053, Australia The right of Mike Begon, Colin Townsend and John Harper to be identified as the Authors of this Work has been asserted in accordance with the UK Copyright, Designs and Patents Act 1988. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by the UK Copyright, Designs, and Patents Act 1988, without the prior permission of the publisher First edition published 1986 by Blackwell Publishing Ltd Second edition published 1990 Third edition published 1996 Fourth edition published 2006 1 2006 Library of Congress Cataloging-in-Publication Data Begon, Michael. Ecology : from individuals to ecosystems / Michael Begon, Colin R. Townsend, John L. Harper.—4th ed. p. cm. Includes bibliographical references and index. ISBN-13: 978-1-4051-1117-1 (hard cover : alk. paper) ISBN-10: 1-4051-1117-8 (hard cover : alk. paper) 1. Ecology. I. Townsend, Colin R. II. Harper, John L. III. Title. QH54.B416 2005 577—dc22 2005004136 A catalogue record for this title is available from the British Library. Set in 9.5/12 Dante MT by Graphicraft Limited, Hong Kong Printed and bound in the United Kingdom by CPI Bath Press The publisher’s policy is to use permanent paper from mills that operate a sustainable forestry policy, and which has been manufatured from pulp processed using acid-free and elementary chlorine-free practices. Furthermore, the publisher ensures that the text paper and cover board used have met acceptable environmental accreditation standards. For further information on Blackwell Publishing, visit our website: www.blackwellpublishing.com EIPA01 10/24/05 1:36 PM Page iv •••• Contents Preface, vii Introduction: Ecology and its Domain, xi Part 1: Organisms 1 Organisms in their Environments: the Evolutionary Backdrop, 3 2 Conditions, 30 3 Resources, 58 4 Life, Death and Life Histories, 89 5 Intraspecific Competition, 132 6 Dispersal, Dormancy and Metapopulations, 163 7 Ecological Applications at the Level of Organisms and Single-Species Populations: Restoration, Biosecurity and Conservation, 186 Part 2: Species Interactions 8 Interspecific Competition, 227 9 The Nature of Predation, 266 10 The Population Dynamics of Predation, 297 11 Decomposers and Detritivores, 326 12 Parasitism and Disease, 347 13 Symbiosis and Mutualism, 381 14 Abundance, 410 15 Ecological Applications at the Level of Population Interactions: Pest Control and Harvest Management, 439 EIPA01 10/24/05 1:36 PM Page v •••• vi CONTENTS Part 3: Communities and Ecosystems 16 The Nature of the Community: Patterns in Space and Time, 469 17 The Flux of Energy through Ecosystems, 499 18 The Flux of Matter through Ecosystems, 525 19 The Influence of Population Interactions on Community Structure, 550 20 Food Webs, 578 21 Patterns in Species Richness, 602 22 Ecological Applications at the Level of Communities and Ecosystems: Management Based on the Theory of Succession, Food Webs, Ecosystem Functioning and Biodiversity, 633 References, 659 Organism Index, 701 Subject Index, 714 Color plate section between pp. 000 and 000 EIPA01 10/24/05 1:36 PM Page vi PREFACE ix collaborated with John over so many years, but also just how much we learnt from him. We cannot promise to have absorbed or, to be frank, to have accepted, every one of his views; and we hope in particular, in this fourth edition, that we have not strayed too far from the paths through which he has guided us. But if readers recognize any attempts to stimulate and inspire rather than simply to inform, to question rather than to accept, to respect our readers rather than to patronize them, and to avoid unques- tioning obedience to current reputation while acknowledging our debt to the masters of the past, then they will have identified John’s intellectual legacy still firmly imprinted on the text. In previous editions we thanked the great many friends and colleagues who helped us by commenting on various drafts of the text. The effects of their contributions are still strongly evident in the present edition. This fourth edition was also read by a series of reviewers, to whom we are deeply grateful. Several remained anonymous and so we cannot thank them by name, but we are delighted to be able to acknowledge the help of Jonathan Anderson, Mike Bonsall, Angela Douglas, Chris Elphick, Valerie Eviner, Andy Foggo, Jerry Franklin, Kevin Gaston, Charles Godfray, Sue Hartley, Marcel Holyoak, Jim Hone, Peter Hudson, Johannes Knops, Xavier Lambin, Svata Louda, Peter Morin, Steve Ormerod, Richard Sibly, Andrew Watkinson, Jacob Weiner, and David Wharton. At Blackwell, and in the production stage, we were particularly helped and encouraged by Jane Andrew, Elizabeth Frank, Rosie Hayden, Delia Sandford and Nancy Whilton. This book is dedicated to our families – by Mike to Linda, Jessica and Robert, and by Colin to Laurel, Dominic, Jenny and Brennan, and especially to the memory of his mother, Jean Evelyn Townsend. Mike Begon Colin Townsend •••• EIPA01 10/24/05 1:36 PM Page ix •••• EIPA01 10/24/05 1:36 PM Page x •••• Definition and scope of ecology The word ‘ecology’ was first used by Ernest Haeckel in 1869. Paraphrasing Haeckel we can describe ecology as the scientific study of the interactions between organisms and their environ- ment. The word is derived from the Greek oikos, meaning ‘home’. Ecology might therefore be thought of as the study of the ‘home life’ of living organisms. A less vague definition was suggested by Krebs (1972): ‘Ecology is the scientific study of the interactions that determine the distribution and abundance of organisms’. Notice that Krebs’ definition does not use the word ‘environment’; to see why, it is necessary to define the word. The environment of an organism consists of all those factors and phenomena outside the organism that influence it, whether these are physical and chemical (abiotic) or other organisms (biotic). The ‘interactions’ in Krebs’ definition are, of course, interactions with these very factors. The environment therefore retains the central position that Haeckel gave it. Krebs’ definition has the merit of pinpointing the ultimate subject matter of ecology: the distribu- tion and abundance of organisms – where organisms occur, how many occur there, and why. This being so, it might be better still to define ecology as: the scientific study of the distribution and abundance of organisms and the interactions that determine distribution and abundance. As far as the subject matter of ecology is concerned, ‘the distribution and abundance of organisms’ is pleasantly succinct. But we need to expand it. The living world can be viewed as a biological hierarchy that starts with subcellular particles, and continues up through cells, tissues and organs. Ecology deals with the next three levels: the individual organism, the population (consisting of individuals of the same species) and the community (consisting of a greater or lesser number of species populations). At the level of the organism, ecology deals with how individuals are affected by (and how they affect) their environment. At the level of the population, ecology is concerned with the presence or absence of particular species, their abundance or rarity, and with the trends and fluctuations in their numbers. Community ecology then deals with the composition and organization of ecological communities. Ecologists also focus on the pathways followed by energy and matter as these move among living and nonliving elements of a further category of organization: the ecosystem, comprising the community together with its physical environment. With this in mind, Likens (1992) would extend our preferred definition of ecology to include ‘the interactions between organisms and the transformation and flux of energy and matter’. However, we take energy/matter transformations as being subsumed in the ‘interactions’ of our definition. There are two broad approaches that ecologists can take at each level of ecological organization. First, much can be gained by building from properties at the level below: physiology when studying organismal ecology; individual clutch size and survival probabilities when investigating the dynamics of individual species populations; food consumption rates when dealing with inter- actions between predator and prey populations; limits to the similarity of coexisting species when researching communities, and so on. An alternative approach deals directly with properties of the level of interest – for example, niche breadth at the organis- mal level; relative importance of density-dependent processes at the population level; species diversity at the level of community; rate of biomass production at the ecosystem level – and tries to relate these to abiotic or biotic aspects of the environment. Both approaches have their uses, and both will be used in each of the three parts of this book: Organisms; Species Interactions; and Communities and Ecosystems. Introduction: Ecology and its Domain EIPA01 10/24/05 1:36 PM Page xi •• EIPC01 10/24/05 1:42 PM Page 2 •• 1.1 Introduction: natural selection and adaptation From our definition of ecology in the Preface, and even from a layman’s understanding of the term, it is clear that at the heart of ecology lies the relationship between organisms and their environments. In this opening chapter we explain how, funda- mentally, this is an evolutionary relationship. The great Russian– American biologist Theodosius Dobzhansky famously said: ‘Nothing in biology makes sense, except in the light of evolution’. This is as true of ecology as of any other aspect of biology. Thus, we try here to explain the processes by which the properties of different sorts of species make their life possible in particular environments, and also to explain their failure to live in other environments. In mapping out this evolutionary backdrop to the subject, we will also be introducing many of the questions that are taken up in detail in later chapters. The phrase that, in everyday speech, is most commonly used to describe the match between organisms and environment is: ‘organism X is adapted to’ followed by a description of where the organism is found. Thus, we often hear that ‘fish are adapted to live in water’, or ‘cacti are adapted to live in conditions of drought’. In everyday speech, this may mean very little: simply that fish have characteristics that allow them to live in water (and perhaps exclude them from other environments) or that cacti have characteristics that allow them to live where water is scarce. The word ‘adapted’ here says nothing about how the characteristics were acquired. For an ecologist or evolutionary biologist, however, ‘X is adapted to live in Y’ means that environment Y has provided forces of natural selection that have affected the life of X’s ancestors and so have molded and specialized the evolution of X. ‘Adaptation’ means that genetic change has occurred. Regrettably, though, the word ‘adaptation’ implies that organisms are matched to their present environments, suggest- ing ‘design’ or even ‘prediction’. But organisms have not been designed for, or fitted to the present: they have been molded (by natural selection) by past environments. Their characteristics reflect the successes and failures of ancestors. They appear to be apt for the environments that they live in at present only because present environments tend to be similar to those of the past. The theory of evolution by natural selection is an ecological theory. It was first elaborated by Charles Darwin (1859), though its essence was also appreciated by a contemporary and corres- pondent of Darwin’s, Alfred Russell Wallace (Figure 1.1). It rests on a series of propositions. 1 The individuals that make up a population of a species are not identical: they vary, although sometimes only slightly, in size, rate of development, response to temperature, and so on. 2 Some, at least, of this variation is heritable. In other words, the characteristics of an individual are determined to some extent by its genetic make-up. Individuals receive their genes from their ancestors and therefore tend to share their characteristics. 3 All populations have the potential to populate the whole earth, and they would do so if each individual survived and each indi- vidual produced its maximum number of descendants. But they do not: many individuals die prior to reproduction, and most (if not all) reproduce at a less than maximal rate. 4 Different ancestors leave different numbers of descendants. This means much more than saying that different individuals produce different numbers of offspring. It includes also the chances of survival of offspring to reproductive age, the survival and reproduction of the progeny of these offspring, the survival and reproduction of their offspring in turn, and so on. 5 Finally, the number of descendants that an individual leaves depends, not entirely but crucially, on the interaction between the characteristics of the individual and its environment. the meaning of adaptation evolution by natural selection Chapter 1 Organisms in their Environments: the Evolutionary Backdrop EIPC01 10/24/05 1:42 PM Page 3 4 CHAPTER 1 In any environment, some individuals will tend to survive and reproduce better, and leave more descendants, than others. If, because of this, the heritable characteristics of a population change from generation to generation, then evolution by nat- ural selection is said to have occurred. This is the sense in which nature may loosely be thought of as selecting. But nature does not select in the way that plant and animal breeders select. Breeders have a defined end in view – bigger seeds or a faster racehorse. But nature does not actively select in this way: it simply sets the scene within which the evolutionary play of differential survival and reproduction is played out. The fittest individuals in a popula- tion are those that leave the greatest number of descendants. In practice, the term is often applied not to a single individual, but to a typ- ical individual or a type. For example, we may say that in sand dunes, yellow-shelled snails are fitter than brown-shelled snails. Fitness, then, is a relative not an absolute term. The fittest indi- viduals in a population are those that leave the greatest number of descendants relative to the number of descendants left by other individuals in the population. When we marvel at the diversity of complex specializations, there is a temptation to regard each case as an example of evolved perfection. But this would be wrong. The evolutionary process works on the genetic variation that is avail- able. It follows that natural selection is unlikely to lead to the evolution of perfect, ‘maximally fit’ individuals. Rather, organisms •••• Figure 1.1 (a) Charles Darwin, 1849 (lithograph by Thomas H. Maguire; courtesy of The Royal Institution, London, UK/Bridgeman Art Library). (b) Alfred Russell Wallace, 1862 (courtesy of the Natural History Museum, London). fitness: it’s all relative evolved perfection? no (a) (b) EIPC01 10/24/05 1:42 PM Page 4 THE EVOLUTIONARY BACKDROP 7 the extent to which the genetic response of populations to environmental change will always be out of step with the environment and unable to anticipate changing circumstances – this is illustrated in the peppered moth example below. Many polymorphisms, however, are actively maintained in a population by natural selection, and there are a num- ber of ways in which this may occur. 1 Heterozygotes may be of superior fitness, but because of the mechanics of Mendelian genetics they continually generate less fit homozygotes within the population. Such ‘heterosis’ is seen in human sickle-cell anaemia where malaria is prevalent. The malaria parasite attacks red blood cells. The sickle-cell muta- tion gives rise to red cells that are physiologically imperfect and misshapen. However, sickle-cell heterozygotes are fittest because they suffer only slightly from anemia and are little affected by malaria; but they continually generate homozygotes that are either dangerously anemic (two sickle-cell genes) or susceptible to malaria (no sickle-cell genes). None the less, the superior fitness of the heterozygote maintains both types of gene in the population (that is, a polymorphism). 2 There may be gradients of selective forces favoring one form (morph) at one end of the gradient, and another form at the other. This can produce polymorphic populations at inter- mediate positions in the gradient – this, too, is illustrated below in the peppered moth study. 3 There may be frequency-dependent selection in which each of the morphs of a species is fittest when it is rarest (Clarke & Partridge, 1988). This is believed to be the case when rare color forms of prey are fit because they go unrecognized and are therefore ignored by their predators. 4 Selective forces may operate in different directions within different patches in the population. A striking example of this is provided by a reciprocal transplant study of white clover (Trifolium repens) in a field in North Wales (UK). To determine whether the characteristics of individuals matched local features of their environment, Turkington and Harper (1979) removed plants from marked positions in the field and multiplied them into clones in the common environment of a greenhouse. They then transplanted samples from each clone into the place in the sward of vegetation from which it had originally been taken (as a control), and also to the places from where all the others had been taken (a transplant). The plants were allowed to grow for a year before they were removed, dried and weighed. The mean weight of clover plants transplanted back into their home sites was 0.89 g but at away sites it was only 0.52 g, a statistically highly significant difference. This provides strong, direct evidence that clover clones in the pasture had evolved to become specialized such that they performed best in their local environment. But all this was going on within a single population, which was therefore polymorphic. In fact, the distinction between local ecotypes and polymorphic popu- lations is not always a clear one. This is illustrated by another study in North Wales, where there was a gradation in habitats at the margin between maritime cliffs and grazed pasture, and a common species, creeping bent grass (Agrostis stolonifera), was present in many of the habitats. Figure 1.4 shows a map of the site and one of the transects from which plants were sampled. It also shows the results when plants from the sampling points along this transect were grown in a common garden. The •••• Figure 1.4 (a) Map of Abraham’s Bosom, the site chosen for a study of evolution over very short distances. The darker colored area is grazed pasture; the lighter areas are the cliffs falling to the sea. The numbers indicate the sites from which the grass Agrostis stolonifera was sampled. Note that the whole area is only 200 m long. (b) A vertical transect across the study area showing the gradual change from pasture to cliff conditions. (c) The mean length of stolons produced in the experimental garden from samples taken from the transect. (From Aston & Bradshaw, 1966.) the maintenance of polymorphisms no clear distinction between local ecotypes and a polymorphism 1 2 3 4 5 N 0 200 m100 Irish Sea (a) 1 2 3 5 4 100 30 20 10 0 E le va tio n (m ) 0 (b) 100 50 25 0 S to lo n le ng th ( cm ) 0 (c) Distance (m) EIPC01 10/24/05 1:42 PM Page 7 8 CHAPTER 1 plants spread by sending out shoots along the ground surface (stolons), and the growth of plants was compared by measuring the lengths of these. In the field, cliff plants formed only short stolons, whereas those of the pasture plants were long. In the experi- mental garden, these differences were maintained, even though the sampling points were typically only around 30 m apart – certainly within the range of pollen dispersal between plants. Indeed, the gradually changing environment along the transect was matched by a gradually changing stolon length, presumably with a genetic basis, since it was apparent in the common garden. Thus, even though the spatial scale was so small, the forces of selection seem to outweigh the mixing forces of hybridization – but it is a moot point whether we should describe this as a small-scale series of local ecotypes or a polymorphic population maintained by a gradient of selection. 1.2.3 Variation within a species with manmade selection pressures It is, perhaps, not surprising that some of the most dramatic examples of local specialization within species (indeed of natural selection in action) have been driven by manmade ecological forces, especially those of environmental pollution. These can provide rapid change under the influence of powerful selection pressures. Industrial melanism, for example, is the phenomenon in which black or blackish forms of species have come to dominate populations in industrial areas. In the dark individuals, a dominant gene is typ- ically responsible for producing an excess of the black pigment melanin. Industrial melanism is known in most industrialized coun- tries and more than 100 species of moth have evolved forms of industrial melanism. •••• f. insularia f. carbonaria f. typica Figure 1.5 Sites in Britain where the frequencies of the pale ( forma typica) and melanic forms of Biston betularia were recorded by Kettlewell and his colleagues. In all more than 20,000 specimens were examined. The principal melanic form ( forma carbonaria) was abundant near industrial areas and where the prevailing westerly winds carry atmospheric pollution to the east. A further melanic form ( forma insularia, which looks like an intermediate form but is due to several different genes controlling darkening) was also present but was hidden where the genes for forma carbonaria were present. (From Ford, 1975.) EIPC01 10/24/05 1:42 PM Page 8 THE EVOLUTIONARY BACKDROP 9 The earliest recorded species to evolve in this way was the peppered moth (Biston betularia); the first black specimen in an otherwise pale popula- tion was caught in Manchester (UK) in 1848. By 1895, about 98% of the Manchester peppered moth popu- lation was melanic. Following many more years of pollution, a large-scale survey of pale and melanic forms of the peppered moth in Britain recorded more than 20,000 specimens between 1952 and 1970 (Figure 1.5). The winds in Britain are predominantly westerlies, spreading industrial pollutants (especially smoke and sulfur dioxide) toward the east. Melanic forms were concentrated toward the east and were completely absent from the unpolluted western parts of England and Wales, northern Scotland and Ireland. Notice from the figure, though, that many populations were polymorphic: melanic and nonmelanic forms coexisted. Thus, the polymorphism seems to be a result both of environ- ments changing (becoming more polluted) – to this extent the poly- morphism is transient – and of there being a gradient of selective pressures from the less polluted west to the more polluted east. The main selective pressure appears to be applied by birds that prey on the moths. In field experiments, large numbers of melanic and pale (‘typical’) moths were reared and released in equal numbers. In a rural and largely unpolluted area of southern England, most of those captured by birds were melanic. In an industrial area near the city of Birmingham, most were typicals (Kettlewell, 1955). Any idea, however, that melanic forms were favored simply because they were camouflaged against smoke- stained backgrounds in the polluted areas (and typicals were favored in unpolluted areas because they were camouflaged against pale backgrounds) may be only part of the story. The moths rest on tree trunks during the day, and nonmelanic moths are well hidden against a background of mosses and lichens. Industrial pollution has not just blackened the moths’ background; sulfur dioxide, especially, has also destroyed most of the moss and lichen on the tree trunks. Thus, sulfur dioxide pollution may have been as important as smoke in selecting melanic moths. In the 1960s, industrialized environments in Western Europe and the United States started to change again, as oil and electricity began to replace coal, and legislation was passed to impose smoke- free zones and to reduce industrial emissions of sulfur dioxide. The frequency of melanic forms then fell back to near pre- Industrial levels with remarkable speed (Figure 1.6). Again, there was transient polymorphism – but this time while populations were en route in the other direction. 1.3 Speciation It is clear, then, that natural selection can force populations of plants and animals to change their character – to evolve. But none of the examples we have considered has involved the evolution of a new species. What, then, justifies naming two populations as different species? And what is the process – ‘speciation’ – by which two or more new species are formed from one original species? 1.3.1 What do we mean by a ‘species’? Cynics have said, with some truth, that a species is what a competent taxonomist regards as a species. On the other hand, back in the 1930s two American biologists, Mayr and Dobzhansky, proposed an empir- ical test that could be used to decide whether two populations were part of the same species or of two different species. They recognized organisms as being members of a single species if they could, at least potentially, breed together in nature to produce fertile offspring. They called a species tested and defined in this way a biological species or biospecies. In the examples that we have used earlier in this chapter we know that melanic and normal peppered moths can mate and that the offspring are fully fertile; this is also true of plants from the different types of Agrostis. They are all variations within species – not separate species. In practice, however, biologists do not apply the Mayr– Dobzhansky test before they recognize every species: there is simply not enough time or resources, and in any case, there are vast portions of the living world – most microorganisms, for example – where an absence of sexual reproduction makes a strict interbreeding criterion inappropriate. What is more important is that the test recognizes a crucial element in the evolutionary process that we have met already in considering specialization •••• industrial melanism in the peppered moth 100 80 60 40 20 0 F re qu en cy 1950 1960 1970 Year 1980 1990 2000 Figure 1.6 Change in the frequency of the carbonaria form of the peppered moth Biston betularia in the Manchester area since 1950. Vertical lines show the standard error and the horizontal lines show the range of years included. (After Cook et al., 1999.) biospecies: the Mayr– Dobzhansky test EIPC01 10/24/05 1:42 PM Page 9 •• 12 CHAPTER 1 •• 14 g 20 g 34 g 21 g 28 g 20 g 13 g 20 g 18 g 21 g 34 g 8 g 13 g 10 g G. fuliginosa G. fortis G. magnirostris G. scandens G. conirostris G. difficilis C. parvulus C. psittacula C. pauper C. pallida P. crassirostris Ce. fusca Pi. inornata Ce. olivacea Scratch for seeds on the ground Feed on seeds on the ground and the flowers and pulp of prickly pear (Opuntia) Feed in trees on beetles Use spines held in the bill to extract insects from bark crevices Feed on leaves, buds and seeds in the canopy of trees Warbler-like birds feeding on small soft insects (b) 10°N 5°N 0° 90°W 85°W 80°W Culpepper Wenman Pinta Galapágos Santa Cruz San Cristobal HoodIsabela Fernandina Cocos Island Pearl Is. (a) Figure 1.9 (a) Map of the Galápagos Islands showing their position relative to Central America; on the equator 5° equals approximately 560 km. (b) A reconstruction of the evolutionary history of the Galápagos finches based on variation in the length of microsatellite deoxyribonucleic acid (DNA). The feeding habits of the various species are also shown. Drawings of the birds are proportional to actual body size. The maximum amount of black coloring in male plumage and the average body mass are shown for each species. The genetic distance (a measure of the genetic difference) between species is shown by the length of the horizontal lines. Notice the great and early separation of the warbler finch (Certhidea olivacea) from the others, suggesting that it may closely resemble the founders that colonized the islands. C, Camarhynchus; Ce, Certhidea; G, Geospiza; P, Platyspiza; Pi, Pinaroloxias. (After Petren et al., 1999.) EIPC01 10/24/05 1:42 PM Page 12 •• THE EVOLUTIONARY BACKDROP 13 psittacula) has become insectivorous, feeding on beetles and other insects in the canopy of trees. A so-called woodpecker finch, Camarhynchus (Cactospiza) pallida, extracts insects from crevices by holding a spine or a twig in its bill, while yet a fur- ther group includes the warbler finch, which flits around actively and collects small insects in the forest canopy and in the air. Isolation – both of the archipelago itself and of individual islands within it – has led to an original evolutionary line radiating into a series of species, each matching its own environment. 1.4 Historical factors Our world has not been constructed by someone taking each species in turn, testing it against each environment, and molding it so that every species finds its perfect place. It is a world in which species live where they do for reasons that are often, at least in part, accidents of history. We illustrate this first by continuing our examination of islands. 1.4.1 Island patterns Many of the species on islands are either subtly or profoundly dif- ferent from those on the nearest comparable area of mainland. Put simply, there are two main reasons for this. 1 The animals and plants on an island are limited to those types having ancestors that managed to disperse there, although the extent of this limitation depends on the isolation of the island and the intrinsic dispersal ability of the animal or plant in question. 2 Because of this isolation, as we saw in the previous section, the rate of evolutionary change on an island may often be fast enough to outweigh the effects of the exchange of genetic material between the island population and related populations elsewhere. Thus, islands contain many species unique to themselves (‘endemics’ – species found in only one area), as well as many differentiated ‘races’ or ‘subspecies’ that are distinguishable from mainland forms. A few individuals that disperse by chance to a habitable island can form the nucleus of an expanding new species. Its character will have been colored by the particular genes that were represented among the colonists – which are unlikely to be a perfect sample of the parent population. What natural selection can do with this founder population is limited by what is in its limited sample of genes (plus occasional rare mutations). Indeed much of the deviation among populations isolated on islands appears to be due to a founder effect – the chance composition of the pool of founder genes puts limits and constraints on what variation there is for natural selection to act upon. The Drosophila fruit-flies of Hawaii provide a further spec- tacular example of species formation on islands. The Hawaiian chain of islands (Figure 1.10) is volcanic in origin, having been formed gradually over the last 40 million years, as the center of the Pacific tectonic plate moved steadily over a ‘hot spot’ in a southeasterly direction (Niihau is the most ancient of the islands, Hawaii itself the most recent). The richness of the Hawaiian Drosophila is spectacular: there are probably about 1500 Drosophila spp. worldwide, but at least 500 of these are found only in the Hawaiian islands. Of particular interest are the 100 or so species of ‘picture-winged’ Droso- phila. The lineages through which these species have evolved can be traced by analyzing the banding patterns on the giant chro- mosomes in the salivary glands of their larvae. The evolutionary tree that emerges is shown in Figure 1.10, with each species lined up above the island on which it is found (there are only two species found on more than one island). The historical element in ‘what lives where’ is plainly apparent: the more ancient species live on the more ancient islands, and, as new islands have been formed, rare dispersers have reached them and eventually evolved in to new species. At least some of these species appear to match the same environment as others on different islands. Of the closely related species, for example, D. adiastola (species 8) is only found on Maui and D. setosimentum (species 11) only on Hawaii, but the environments that they live in are apparently indistinguishable (Heed, 1968). What is most noteworthy, of course, is the power and importance of isolation (coupled with natural selection) in generating new species. Thus, island biotas illustrate two import- ant, related points: (i) that there is a historical element in the match between organisms and environments; and (ii) that there is not just one perfect organism for each type of environment. 1.4.2 Movements of land masses Long ago, the curious distributions of species between continents, seemingly inexplicable in terms of dispersal over vast distances, led biologists, especially Wegener (1915), to suggest that the continents themselves must have moved. This was vigorously denied by geologists, until geomagnetic measurements required the same, apparently wildly improbable explanation. The discovery that the tectonic plates of the earth’s crust move and carry with them the migrating continents, reconciles geologist and biologist (Figure 1.11b–e). Thus, whilst major evolutionary developments were occurring in the plant and animal kingdoms, populations were being split and separated, and land areas were moving across climatic zones. Figure 1.12 shows just one example of a major group of organisms (the large flightless birds), whose distributions begin to make sense only in the light of the movement of land masses. It would be •• Hawaiian Drosophila large flightless birds EIPC01 10/24/05 1:42 PM Page 13 •••• 14 CHAPTER 1 N 62 95 68 70 54 53 43 55 85 86 76 99 81 91 77 84 89 75 59 60 61 67 74 69 83 82 97 90 94 81 50 52 49 51 48 3735 36 38 39 47 44 46 66 58 8180 98 punalua group (58–65) glabriapex group (34–57) grimshawi group (66–101) planitidia group (17–33) 40 41 42 2221 2524 26 27 23 18 19 17 20 34 32 1613 1514 6 4 5 1 adiastola group (3–16) 2 3 Niihau Kauai Oahu Lanai Molokai Maui Kahoolawe Hawaii 63 64 65 71 72 73 78 79 87 88 92 93 96 100 101 57 56 45 33 31 30 29 28 10 8 97 12 11 0 50 km Figure 1.10 An evolutionary tree linking the picture-winged Drosophila of Hawaii, traced by the analysis of chromosomal banding patterns. The most ancient species are D. primaeva (species 1) and D. attigua (species 2), found only on the island of Kauai. Other species are represented by solid circles; hypothetical species, needed to link the present day ones, are represented by open circles. Each species has been placed above the island or islands on which it is found (although Molokai, Lanai and Maui are grouped together). Niihau and Kahoolawe support no Drosophila. (After Carson & Kaneshiro, 1976; Williamson, 1981.) EIPC01 10/24/05 1:42 PM Page 14 THE EVOLUTIONARY BACKDROP 17 climate during the Pleistocene ice ages, in particular, bear a lot of the responsibility for the present patterns of distribution of plants and animals. The extent of these climatic and biotic changes is only beginning to be unraveled as the technology for discover- ing, analyzing and dating biological remains becomes more sophisticated (particularly by the analysis of buried pollen sam- ples). These methods increasingly allow us to determine just how much of the present distribution of organisms represents a precise local match to present environments, and how much is a fingerprint left by the hand of history. Techniques for the measurement of oxygen isotopes in ocean cores indic- ate that there may have been as many as 16 glacial cycles in the Pleistocene, each lasting for about 125,000 years (Figure 1.13a). It seems that each glacial phase may have lasted for as long as 50,000–100,000 years, with brief intervals of 10,000–20,000 years when the tem- peratures rose close to those we experience today. This suggests that it is present floras and faunas that are unusual, because they have developed towards the end of one of a series of unusual catas- trophic warm events! During the 20,000 years since the peak of the last glaciation, global temperatures have risen by about 8°C, and the rate at which vegetation has changed over much of this period has been detected by examining pollen records. The woody species that dominate pollen profiles at Rogers Lake in Connecticut (Figure 1.13b) have arrived in turn: spruce first and chestnut most recently. Each new arrival has added to the number of the species present, which has increased continually over the past 14,000-year period. The same picture is repeated in European profiles. As the number of pollen records has increased, it has become possible not only to plot the changes in vegetation •••• Te m pe ra tu re ( °C ) Time (103 years ago) 30 0 50 150 200 250 (a) 20 100 300 350 400 0 (b) 2 4 6 8 10 12 14 0 0 0 0 0 10,000 0 0 500 0 0 0 500 2000 1000 3000 10,000 20,000 2000 4000 1000 2000 5000 15,000 500 2000 1000 1000 2000 1000 Chestnut Hickory Beech Hemlock Oak Pine Pine Spruce Pi ce a Sp ru ce Pi nu s Pi ne Be tu la Bi rc h Ts ug a He m lo ck Q ue rc us O ak Ac er s ac ch ar um Su ga r m ap le Ac er ru br um Re d m ap le Fa gu s Be ec h Ca ry a Hi ck or y Ca st an ea Ch es tn ut 10 3 ye ar s ag o Figure 1.13 (a) An estimate of the temperature variations with time during glacial cycles over the past 400,000 years. The estimates were obtained by comparing oxygen isotope ratios in fossils taken from ocean cores in the Caribbean. The dashed line corresponds to the ratio 10,000 years ago, at the start of the present warming period. Periods as warm as the present have been rare events, and the climate during most of the past 400,000 years has been glacial. (After Emiliani, 1966; Davis, 1976.) (b) The profiles of pollen accumulated from late glacial times to the present in the sediments of Rogers Lake, Connecticut. The estimated date of arrival of each species in Connecticut is shown by arrows at the right of the figure. The horizontal scales represent pollen influx: 103 grains cm−2 year−1. (After Davis et al., 1973.) the Pleistocene glacial cycles . . . . . . from which trees are still recovering EIPC01 10/24/05 1:42 PM Page 17 18 CHAPTER 1 at a point in space, but to begin to map the movements of the various species as they have spread across the continents (see Bennet, 1986). In the invasions that followed the retreat of the ice in eastern North America, spruce was followed by jack pine or red pine, which spread northwards at a rate of 350–500 m year−1 for several thousands of years. White pine started its migration about 1000 years later, at the same time as oak. Hemlock was also one of the rapid invaders (200–300 m year−1), and arrived at most sites about 1000 years after white pine. Chestnut moved slowly (100 m year−1), but became a dominant species once it had arrived. Forest trees are still migrating into deglaciated areas, even now. This clearly implies that the timespan of an average interglacial period is too short for the attainment of floristic equilibrium (Davis, 1976). Such historical factors will have to be borne in mind when we consider the various patterns in species richness and biodiversity in Chapter 21. ‘History’ may also have an impact on much smaller space and time scales. Disturbances to the benthic (bottom dwelling) community of a stream occurs when high discharge events (associated with storms or snow melt) result in a very small-scale mosaic of patches of scour (substrate loss), fill (addition of substrate) and no change (Matthaei et al., 1999). The invertebrate communities associated with the differ- ent patch histories are distinctive for a period of months, within which time another high discharge event is likely to occur. As with the distribution of trees in relation to repeating ice ages, the stream fauna may rarely achieve an equilibrium between flow disturbances (Matthaei & Townsend, 2000). The records of climatic change in the tropics are far less complete than those for temperate regions. There is therefore the temptation to imagine that whilst dramatic climatic shifts and ice invasions were dom- inating temperate regions, the tropics persisted in the state we know today. This is almost certainly wrong. Data from a variety of sources indicate that there were abrupt fluctuations in post- glacial climates in Asia and Africa. In continental monsoon areas (e.g. Tibet, Ethiopia, western Sahara and subequatorial Africa) the postglacial period started with an extensive phase of high humid- ity followed by a series of phases of intense aridity (Zahn, 1994). In South America, a picture is emerging of vegetational changes that parallel those occurring in temperate regions, as the extent of tropical forest increased in warmer, wetter periods, and con- tracted, during cooler, drier glacial periods, to smaller patches surrounded by a sea of savanna. Support for this comes from the present-day distribution of species in the tropical forests of South America (Figure 1.14). There, particular ‘hot spots’ of species diversity are apparent, and these are thought to be likely sites of forest refuges during the glacial periods, and sites too, there- fore, of increased rates of speciation (Prance, 1987; Ridley, 1993). On this interpretation, the present distributions of species may again be seen as largely accidents of history (where the refuges were) rather than precise matches between species and their dif- fering environments. Evidence of changes in vegetation that followed the last retreat of the ice hint at the consequence of the global warming (maybe 3°C in the next 100 years) that is predicted to result from continuing increases in atmospheric carbon dioxide (discussed in detail in Sections 2.9.1 and 18.4.6). But the scales are quite different. Postglacial warming of about 8°C occurred over 20,000 years, and changes in the vegetation failed to keep pace even with this. But current projections for the 21st century require range shifts for trees at rates of 300–500 km per century compared to typical rates in the past of 20–40 km per century (and exceptional rates of 100–150 km). It is striking that the only pre- cisely dated extinction of a tree species in the Quaternary, that of Picea critchfeldii, occurred around 15,000 years ago at a time of especially rapid postglacial warming ( Jackson & Weng, 1999). Clearly, even more rapid change in the future could result in extinc- tions of many additional species (Davis & Shaw, 2001). •••• Napo Madiera Peru East Imeri Guiana (b)(a) Figure 1.14 (a) The present-day distribution of tropical forest in South America. (b) The possible distribution of tropical forest refuges at the time when the last glaciation was at its peak, as judged by present-day hot spots of species diversity within the forest. (After Ridley, 1993.) ‘history’ on a smaller scale changes in the tropics how will global warming compare? EIPC01 10/24/05 1:42 PM Page 18 THE EVOLUTIONARY BACKDROP 19 1.4.4 Convergents and parallels A match between the nature of organ- isms and their environment can often be seen as a similarity in form and behavior between organisms living in a similar environment, but belonging to different phyletic lines (i.e. different branches of the evolutionary tree). Such similarities also undermine further the idea that for every environment there is one, and only one, perfect organism. The evidence is particularly persuasive when the phyletic lines are far removed from each other, and when similar roles are played by structures that have quite different evolutionary origins, i.e. when the structures are analogous (similar in superficial form or function) but not homologous (derived from an equivalent structure in a common ancestry). When this is seen to occur, we speak of convergent evolution. Many flowering plants and some ferns, for example, use the support of others to climb high in the canopies of vegetation, and so gain access to more light than if they depended on their own supporting tissues. The ability to climb has evolved in many dif- ferent families, and quite different organs have become modified into climbing structures (Figure 1.15a): they are analogous struc- tures but not homologous. In other plant species the same organ has been modified into quite different structures with quite dif- ferent roles: they are therefore homologous, although they may not be analogous (Figure 1.15b). Other examples can be used to show the parallels in evolutionary pathways within separate groups that have radiated after they were isolated from each other. The classic example of such parallel evolution is the radiation amongst the placental and marsupial mammals. Marsupials arrived on the Australian continent in the Cretaceous period (around 90 million years ago), when the only other mammals present were the curious egg-laying monotremes (now represented only by the spiny anteaters (Tachyglossus aculeatus) and the duckbill platypus (Ornithorynchus anatinus)). An evolutionary process of radiation then occurred that in many •••• Dioscorea (Dioscoreaceae), twiner Calamus (Arecaceae), hooks Clematis (Ranunculaceae), twining petiole Cobaea (Cobaeaceae), tendril Ficus (Moraceae), adventitious roots Parthenocissus (Vitaceae), sticky pads (a) analogous and homologous structures Figure 1.15 A variety of morphological features that allow flowering plants to climb. (a) Structural features that are analogous, i.e. derived from modifications of quite different organs, e.g. leaves, petioles, stems, roots and tendrils. EIPC01 10/24/05 1:42 PM Page 19 22 CHAPTER 1 from the Low Arctic (around 600 species in North America) to the High Arctic (north of 83°, e.g. around 100 species in Greenland and Ellesmere Island). In contrast, the flora of Antarctica contains only two native species of vascular plant and some lichens and mosses that support a few small invertebrates. The biological productivity and diversity of Antarctica are con- centrated at the coast and depend almost entirely on resources harvested from the sea. Taiga or northern coniferous forest (see Plate 1.2, facing p. XX) occupies a broad belt across North America and Eurasia. Liquid water is unavailable for much of the winter, and plants and many of the animals have a conspicuous winter dor- mancy in which metabolism is very slow. Generally, the tree flora is very limited. In areas with less severe winters, the forests may be dominated by pines (Pinus species, which are all evergreens) and deciduous trees such as larch (Larix), birch (Betula) or aspens (Populus), often as mixtures of species. Farther north, these species give way to single-species forests of spruce (Picea) cover- ing immense areas. The overriding environmental constraint in northern spruce forests is the presence of permafrost, creating drought except when the sun warms the surface. The root system of spruce can develop in the superficial soil layer, from which the trees derive all their water during the short growing season. Temperate forests (see Plate 1.3, between pp. XX and XX) range from the mixed conifer and broad-leaved forests of much of North America and northern central Europe (where there may be 6 months of freezing temperatures), to the moist dripping forests of broad-leaved evergreen trees found at the biome’s low latitude limits in, for example, Florida and New Zealand. In most temperate forests, however, there are periods of the year when liquid water is in short supply, because poten- tial evaporation exceeds the sum of precipitation and water available from the soil. Deciduous trees, which dominate in most temperate forests, lose their leaves in the fall and become dormant. On the forest floor, diverse floras of perennial herbs often occur, particularly those that grow quickly in the spring before the new tree foliage has developed. Temperate forests also •••• Arctic tundra Northern coniferous forest Temperate forest Tropical rainforest Tropical seasonal forest Temperate grassland Tropical savanna grassland and scrub Desert Mediterranean vegetation, chaparral Mountains Figure 1.17 World distribution of the major biomes of vegetation. (After Audesirk & Audesirk, 1996.) taiga temperate forests EIPC01 10/24/05 1:42 PM Page 22 THE EVOLUTIONARY BACKDROP 23 provide food resources for animals that are usually very seasonal in their occurrence. Many of the birds of temperate forests are migrants that return in spring but spend the remainder of the year in warmer biomes. Grassland occupies the drier parts of temperate and tropical regions. Temperate grassland has many local names: the steppes of Asia, the prairies of North America, the pampas of South America and the veldt of South Africa. Tropical grassland or savanna (see Plate 1.4, between pp. XX and XX) is the name applied to tropical vegetation ranging from pure grass- land to some trees with much grass. Almost all of these temper- ate and tropical grasslands experience seasonal drought, but the role of climate in determining their vegetation is almost completely overridden by the effects of grazing animals that limit the species present to those that can recover from frequent defoliation. In the savanna, fire is also a common hazard in the dry season and, like grazing animals, it tips the balance in the vegetation against trees and towards grassland. None the less, there is typically a sea- sonal glut of food, alternating with shortage, and as a consequence the larger grazing animals suffer extreme famine (and mortality) in drier years. A seasonal abundance of seeds and insects supports large populations of migrating birds, but only a few species can find sufficiently reliable resources to be resident year-round. Many of these natural grasslands have been cultivated and replaced by arable annual ‘grasslands’ of wheat, oats, barley, rye and corn. Such annual grasses of temperate regions, together with rice in the tropics, provide the staple food of human popu- lations worldwide. At the drier margins of the biome, many of the grasslands are ‘managed’ for meat or milk production, some- times requiring a nomadic human lifestyle. The natural popula- tions of grazing animals have been driven back in favor of cattle, sheep and goats. Of all the biomes, this is the one most coveted, used and transformed by humans. Chaparral or maquis occurs in Mediterranean-type climates (mild, wet winters and summer drought) in Europe, California and northwest Mexico, and in a few small areas in Australia, Chile and South Africa. Chaparral develops in regions with less rainfall than temperate grasslands and is dominated mainly by a •••• 40 –60 0 5000 40 –60 0 5000 (b) Savanna 40 –60 0 M in im um te m pe ra tu re (m on th ly a ve ra ge ,° C ) M in im um te m pe ra tu re (m on th ly a ve ra ge ,° C ) 5000 (c) Temperate deciduous forest (d) Northern coniferous forest (taiga) (e) Tundra Total annual rainfall (mm) (a) Tropical rainforest 40 –60 0 5000 40 –60 0 5000 M in im um te m pe ra tu re (m on th ly a ve ra ge ,° C ) Congo (Africa) Manaus (South America) Atherton (Australia) Figure 1.18 The variety of environmental conditions experienced in terrestrial environments can be described in terms of their annual rainfall and mean monthly minimum temperatures. The range of conditions experienced in: (a) tropical rainforest, (b) savanna, (c) temperate deciduous forest, (d) northern coniferous forest (taiga), and (e) tundra. (After Heal et al., 1993; © UNESCO.) grassland chaparral EIPC01 10/24/05 1:42 PM Page 23 24 CHAPTER 1 drought-resistant, hard-leaved scrub of low-growing woody plants. Annual plants are also common in chaparral regions dur- ing the winter and early spring, when rainfall is more abundant. Chaparral is subject to periodic fires; many plants produce seeds that will only germinate after fire while others can quickly resprout because of food reserves in their fire-resistant roots. Deserts (see Plate 1.5, between pp. XX and XX) are found in areas that experi- ence extreme water shortage: rainfall is usually less than about 25 cm year−1, is usually very unpredictable and is considerably less than potential evaporation. The desert biome spans a very wide range of temperatures, from hot deserts, such as the Sahara, to very cold deserts, such as the Gobi in Mongolia. In their most extreme form, the hot deserts are too arid to bear any vegetation; they are as bare as the cold deserts of Antarctica. Where there is sufficient rainfall to allow plants to grow in arid deserts, its timing is always unpredictable. Desert vegetation falls into two sharply contrasted patterns of behavior. Many species have an opportunistic lifestyle, stimulated into germination by the unpredictable rains. They grow fast and complete their life history by starting to set new seed after a few weeks. These are the species that can occasionally make a desert bloom. A different pattern of behavior is to be long-lived with sluggish physiological processes. Cacti and other succulents, and small shrubby species with small, thick and often hairy leaves, can close their stomata (pores through which gas exchange takes place) and tolerate long periods of physiological inactivity. The relative poverty of animal life in arid deserts reflects the low productiv- ity of the vegetation and the indigestibility of much of it. Tropical rainforest (see Plate 1.6, between pp. XX and XX) is the most productive of the earth’s biomes – a result of the coincidence of high solar radiation received through- out the year and regular and reliable rainfall. The productivity is achieved, overwhelmingly, high in the dense forest canopy of evergreen foliage. It is dark at ground level except where fallen trees create gaps. Often, many tree seedlings and saplings remain in a suppressed state from year to year and only leap into action if a gap forms in the canopy above them. Apart from the trees, the vegetation is largely composed of plant forms that reach up into the canopy vicariously; they either climb and then scramble in the tree canopy (vines and lianas, including many species of fig) or grow as epiphytes, rooted on the damp upper branches. Most species of both animals and plants in tropical rain forest are active throughout the year, though the plants may flower and ripen fruit in sequence. Dramatically high species richness is the norm for tropical rainforest, and communities rarely if ever become dom- inated by one or a few species. The diversity of rainforest trees provides for a corresponding diversity of resources for herbivores, and so on up the food chain. Erwin (1982) estimated that there are 18,000 species of beetle in 1 ha of Panamanian rainforest (compared with only 24,000 in the whole of the United States and Canada!). All of these biomes are terrestrial. Aquatic ecologists could also come up with a set of biomes, although the tra- dition has largely been a terrestrial one. We might distinguish springs, rivers, ponds, lakes, estuaries, coastal zones, coral reefs and deep oceans, among other distinctive kinds of aquatic com- munity. For present purposes, we recognize just two aquatic biomes, marine and freshwater. The oceans cover about 71% of the earth’s surface and reach depths of more than 10,000 m. They extend from regions where precipitation exceeds evapora- tion to regions where the opposite is true. There are massive move- ments within this body of water that prevent major differences in salt concentrations developing (the average concentration is about 3%). Two main factors influence the biological activity of the oceans. Photosynthetically active radiation is absorbed in its pas- sage through water, so photosynthesis is confined to the surface region. Mineral nutrients, especially nitrogen and phosphorus, are commonly so dilute that they limit the biomass that can develop. Shallow waters (e.g. coastal regions and estuaries) tend to have high biological activity because they receive mineral input from the land and less incident radiation is lost than in passage through deep waters. Intense biological activity also occurs where nutrient-rich waters from the ocean depths come to the surface; this accounts for the concentration of many of the world’s fisheries in Arctic and Antarctic waters. Freshwater biomes occur mainly on the route from land drainage to the sea. The chemical composition of the water varies enormously, depending on its source, its rate of flow and the inputs of organic matter from vegetation that is rooted in or around the aquatic environment. In water catchments where the rate of evaporation is high, salts leached from the land may accumulate and the concentrations may far exceed those present in the oceans; brine lakes or even salt pans may be formed in which little life is possible. Even in aquatic situations liquid water may be unavailable, as is the case in the polar regions. Differentiating between biomes allows only a very crude recognition of the sorts of differences and similarities that occur between communities of organisms. Within biomes there are both small- and large-scale patterns of variation in the structure of com- munities and in the organisms that inhabit them. Moreover, as we see next, what characterizes a biome is not necessarily the particular species that live there. 1.5.2 The ‘life form spectra’ of communities We pointed out earlier the crucial importance of geographic isolation in allowing populations to diverge under selection. The geographic distributions of species, genera, families and even higher taxonomic categories of plants and animals often reflect this geographic divergence. All species of lemurs, for example, are found on the island of Madagascar and nowhere else. Similarly, •••• desert tropical rainforest aquatic biomes? EIPC01 10/24/05 1:42 PM Page 24 THE EVOLUTIONARY BACKDROP 27 daily and seasonal cycles), directional (like the accumulation of a pollutant in a lake) or erratic (like fires, hailstorms and typhoons). Heterogeneity crops up again and again in later chapters – in part because of the challenges it poses to organisms in moving from patch to patch (Chapter 6), in part because of the variety of opportunities it provides for different species (Chapters 8 and 19), and in part because heterogeneity can alter communities by interrupting what would otherwise be a steady march to an equilibrium state (Chapters 10 and 19). 1.6.2 Pairs of species As we have already noted, the existence of one type of organism in an area immediately diversifies it for others. Over its lifetime, an organism may increase the diversity of its environment by con- tributing dung, urine, dead leaves and ultimately its dead body. During its life, its body may serve as a place in which other species find homes. Indeed, some of the most strongly developed matches between organisms and their environment are those in which one species has developed a dependence upon another. This is the case in many relationships between consumers and their foods. Whole syndromes of form, behavior and metabolism constrain the animal within its narrow food niche, and deny it access to what might otherwise appear suitable alternative foods. Similar tight matches are characteristic of the relationships between parasites and their hosts. The various interactions in which one species is consumed by another are the subject matter of Chapters 9–12. Where two species have evolved a mutual dependence, the fit may be even tighter. We examine such ‘mutualisms’ in detail in Chapter 13. The association of nitrogen-fixing bacteria with the roots of leguminous plants, and the often extremely precise rela- tionships between insect pollinators and their flowers, are two good examples. When a population has been exposed to variations in the phys- ical factors of the environment, for example a short growing season or a high risk of frost or drought, a once-and-for-all toler- ance may ultimately evolve. The physical factor cannot itself change or evolve as a result of the evolution of the organisms. By contrast, when members of two species interact, the change in each produces alterations in the life of the other, and each may generate selective forces that direct the evolution of the other. In such a coevolutionary process the interaction between two species may continually escalate. What we then see in nature may be pairs of species that have driven each other into ever narrowing ruts of specialization – an ever closer match. •••• N um be r of s pe ci es ( % ) CI 0 HF 20 30 40 M (a) 10 CI 0 HF 20 30 40 M (b) 10 N um be r of s pe ci es ( % ) CI 0 HF 20 30 40 M (c) 10 CI 0 HF 20 30 40 M (d) 10 Figure 1.20 The percentages of forest mammals in various locomotory and feeding habitat categories in communities in: (a) Malaya, all forested areas (161 species), (b) Panama dry forest (70 species), (c) Australia, Cape York forest (50 species), and (d) Zaire, Irangi forest (96 species). C, carnivores; HF, herbivores and fructivores; I, insectivores; M, mixed feeders; ( ) aerial; ( ) arboreal; ( ) scansorial; ( ) small ground mammals. (After Andrews et al., 1979.) EIPC01 10/24/05 1:42 PM Page 27 28 CHAPTER 1 1.6.3 Coexistence of similar species While it is no surprise that species with rather different roles coex- ist within the same community, it is also generally the case that communities support a variety of species performing apparently rather similar roles. The Antarctic seals are an example. It is thought that the ancestral seals evolved in the northern hemisphere, where they are present as Miocene fossils, but one group of seals moved south into warmer waters and probably colonized the Antarctic in the Late Miocene or Early Pliocene (about 5 million years ago). When they entered the Antarctic, the Southern Ocean was probably rich in food and free from major predators, as it is today. It was within this environment that the group appears to have undergone radiative evolution (Figure 1.21). For ex- ample, the Weddell seal feeds primarily on fish and has unspe- cialized dentition; the crab-eater seal feeds almost exclusively on krill and its teeth are suited to filtering these from the sea water; the Ross seal has small, sharp teeth and feeds mainly on pelagic squid; and the leopard seal has large, cusped, grasping teeth and feeds on a wide variety of foods, including other seals and, in some seasons, penguins. Do these species compete with one another? Do competing species need to be different if they are to coexist? If so, how dif- ferent do they need to be: is there some limit to their similarity? Do species like the seals interact with one another at the present time, or has evolution in the past led to the absence of such interactions in contemporary communities? We return to these questions about coexisting, similar species in Chapter 8. Even at this stage, though, we may note that coexisting species, even when apparently very similar, commonly differ in subtle ways – not simply in their morphology or physiology but also in their responses to their environment and the role they play within the community of which they are part. The ‘ecological niches’ of such species are said to be differentiated from one another. The concept of the ecological niche is itself explained in the next two chapters. •••• (c) (a) (d) (b) Figure 1.21 Antarctic seals, similar species that coexist: (a) the Weddell seal, Leptonychotes weddellii (© Imageshop – zefa visual media uk ltd/Alamy), (b) the crab-eater seal Lobodon carcinophagus (© Bryan & Cherry Alexander Photography/Alamy), (c) the Ross seal, Omatophoca rossii (© Chris Sattlberger/Science Photo Library), and (d) the leopard seal, Hydrurga leptonyx (© Kevin Schafer/Alamy). EIPC01 10/24/05 1:42 PM Page 28 THE EVOLUTIONARY BACKDROP 29 Summary ‘Nothing in biology makes sense, except in the light of evolution’. We try in this chapter to illustrate the processes by which the properties of different sorts of species make their life possible in particular environments. We explain what is meant by evolutionary adaptation and by the theory of evolution by natural selection, an ecological theory first elaborated by Charles Darwin in 1859. Through natural selection, organisms come to match their environments by being ‘the fittest available’ or ‘the fittest yet’: they are not ‘the best imaginable’. Adaptive variation within species can occur at a range of levels: all represent a balance between local adaptation and hybridization. Ecotypes are genetically determined variants between populations within a species that reflect local matches between the organisms and their environments. Genetic poly- morphism is the occurrence together in the same habitat of two or more distinct forms. Dramatic examples of local specialization have been driven by manmade ecological forces, especially those of environmental pollution. We describe the process of speciation by which two or more new species are formed from one original species and explain what we mean by a ‘species’, especially a biospecies. Islands provide arguably the most favorable environment for populations to diverge into distinct species. Species live where they do for reasons that are often accidents of history. We illustrate this by examining island patterns, the move- ments of land masses over geological time, climatic changes especially during the Pleistocene ice ages (and we compare this with predicted changes consequent on current global warming) and the concepts of convergent and parallel evolution. The various terrestrial biomes of the earth are reviewed and their aquatic equivalents touched on briefly. Raunkiaer’s concept of life form spectra, in particular, emphasizes that ecological communities may be fundamentally very similar even when taxonomically quite distinct. All communities comprise a diversity of species: a diversity of matches to the local environment. Environmental heterogeneity, interactions between predators and prey, parasites and hosts and mutualists, and the coexistence of similar species all contribute to this. •••• EIPC01 10/24/05 1:42 PM Page 29 32 CHAPTER 2 2.3 Responses of individuals to temperature 2.3.1 What do we mean by ‘extreme’? It seems natural to describe certain environmental conditions as ‘extreme’, ‘harsh’, ‘benign’ or ‘stressful’. It may seem obvious when conditions are ‘extreme’: the midday heat of a desert, the cold of an Antarctic winter, the salinity of the Great Salt Lake. But this only means that these conditions are extreme for us, given our particular physiological characteristics and tolerances. To a cactus there is nothing extreme about the desert condi- tions in which cacti have evolved; nor are the icy fastnesses of Antarctica an extreme environment for penguins (Wharton, 2002). It is too easy and dangerous for the ecologist to assume that all other organisms sense the environment in the way we do. Rather, the ecologist should try to gain a worm’s-eye or plant’s-eye view of the environment: to see the world as others see it. Emotive words like harsh and benign, even relat- ivities such as hot and cold, should be used by ecologists only with care. •••• Ranunculus glacialis Oxyria digyna Geum reptans Pinus cembra Picea abies Betula pendula Larix decidua Picea abies Larix decidua Leucojum vernum Betula pendula Fagus sylvatica Taxus baccata Abies alba Prunus laurocerasus Quercus ilex Olea europaea Quercus pubescens Citrus limonum Te m pe ra tu re ( °C ) 25 20 15 10 5 Salinity (%) 0 10 20 30 405 15 25 35 45 2600 2500 2500 1900 1900 1900 1900 900 900 600 600 600 550 530 250 240 240 240 80 (m) (a) (b)Temperature (°C) 5 10 15 20 25 30 100% mortality 50% mortality Zero mortality Temperature pH (c) Fo od a va ila ble Figure 2.2 (a) A niche in one dimension. The range of temperatures at which a variety of plant species from the European Alps can achieve net photosynthesis of low intensities of radiation (70 W m−2). (After Pisek et al., 1973.) (b) A niche in two dimensions for the sand shrimp (Crangon septemspinosa) showing the fate of egg-bearing females in aerated water at a range of temperatures and salinities. (After Haefner, 1970.) (c) A diagrammatic niche in three dimensions for an aquatic organism showing a volume defined by the temperature, pH and availability of food. EIPC02 10/24/05 1:44 PM Page 32 CONDITIONS 33 2.3.2 Metabolism, growth, development and size Individuals respond to temperature essentially in the manner shown in Figure 2.1a: impaired function and ultimately death at the upper and lower extremes (discussed in Sec- tions 2.3.4 and 2.3.6), with a functional range between the extremes, within which there is an optimum. This is accounted for, in part, simply by changes in metabolic effectiveness. For each 10°C rise in temperature, for example, the rate of biological enzy- matic processes often roughly doubles, and thus appears as an exponential curve on a plot of rate against temperature (Figure 2.3). The increase is brought about because high temperature increases the speed of molecular movement and speeds up chemical reac- tions. The factor by which a reaction changes over a 10°C range is referred to as a Q10: a rough doubling means that Q10 ≈ 2. For an ecologist, however, effects on individual chemical reactions are likely to be less important than effects on rates of growth (increases in mass), on rates of development (progression through lifecycle stages) and on final body size, since, as we shall discuss much more fully in Chapter 4, these tend to drive the core ecological activities of survival, reproduction and movement. And when we plot rates of growth and development of whole organisms against temperature, there is quite com- monly an extended range over which there are, at most, only slight deviations from linearity (Figure 2.4). When the relationship between growth or development is effectively linear, the temperatures experienced by an organism can be summarized in a single very useful value, the number of ‘day- degrees’. For instance, Figure 2.4c shows that at 15°C (5.1°C above a development threshold of 9.9°C) the predatory mite, Amblyseius californicus, took 24.22 days to develop (i.e. the proportion of its total development achieved each day was 0.041 (= 1/24.22)), but it took only 8.18 days to develop at 25°C (15.1°C above the same threshold). At both temperatures, therefore, development required 123.5 day-degrees (or, more properly, ‘day-degrees above thresh- old’), i.e. 24.22 × 5.1 = 123.5, and 8.18 × 15.1 = 123.5. This is also the requirement for development in the mite at other temper- atures within the nonlethal range. Such organisms cannot be said to require a certain length of time for development. What they require is a combination of time and temperature, often referred to as ‘physiological time’. Together, the rates of growth and development determine the final size of an organism. For instance, for a given rate of growth, a faster rate of devel- opment will lead to smaller final size. Hence, if the responses of growth and development to variations in temperature are not the same, temperature will also affect final size. In fact, development usually increases more rapidly with temperature than does growth, such that, for a very wide range of organisms, final size tends to decrease with rearing temperature: the ‘temperature–size rule’ (see Atkinson et al., 2003). An example for single-celled protists (72 data sets from marine, brackish and freshwater habitats) is shown in Figure 2.5: for each 1°C increase in temperature, final cell volume decreased by roughly 2.5%. These effects of temperature on growth, development and size may be of practical rather than simply scientific importance. Increasingly, ecologists are called upon to predict. We may wish to know what the consequences would be, say, of a 2°C rise in temperature resulting from global warming (see Section 2.9.2). Or we may wish to understand the role of temperature in sea- sonal, interannual and geographic variations in the productivity of, for example, marine ecosystems (Blackford et al., 2004). We cannot afford to assume exponential relationships with temper- ature if they are really linear, nor to ignore the effects of changes in organism size on their role in ecological communities. Motivated, perhaps, by this need to be able to extrapolate from the known to the unknown, and also simply by a wish to discover fundamental organiz- ing principles governing the world •••• exponential effects of temperature on metabolic reactions effectively linear effects on rates of growth and development Temperature (°C) 5 10 15 20 25 30 O xy ge n co ns um pt io n (µ l O 2 g– 1 h– 1 ) 600 500 400 300 200 100 Figure 2.3 The rate of oxygen consumption of the Colorado beetle (Leptinotarsa decemineata), which doubles for every 10°C rise in temperature up to 20°C, but increases less fast at higher temperatures. (After Marzusch, 1952.) day-degree concept temperature–size rule ‘universal temperature dependence’? EIPC02 10/24/05 1:44 PM Page 33 •• 34 CHAPTER 2 around us, there have been attempts to uncover universal rules of temperature dependence, for metabolism itself and for develop- ment rates, linking all organisms by scaling such dependences with aspects of body size (Gillooly et al., 2001, 2002). Others have suggested that such generalizations may be oversimplified, stress- ing for example that characteristics of whole organisms, like growth and development rates, are determined not only by the temperature dependence of individual chemical reactions, but also by those of the availability of resources, their rate of diffusion from the environment to metabolizing tissues, and so on (Rombough, 2003; Clarke, 2004). It may be that there is room for coexistence between broad-sweep generalizations at the grand scale and the more complex relationships at the level of individual species that these generalizations subsume. 2.3.3 Ectotherms and endotherms Many organisms have a body temperature that differs little, if at all, from their environment. A parasitic worm in the gut of a mammal, a fungal mycelium in the soil and a sponge in the sea acquire the temperature of the medium in which they live. Terrestrial organisms, exposed to the sun and the air, are differ- ent because they may acquire heat directly by absorbing solar radi- ation or be cooled by the latent heat of evaporation of water (typical •• G ro w th r at e (µ m d ay –1 ) –0.2 4 1.0 Temperature (°C) 0.8 246 8 10 12 14 16 18 20 22 (a) 0.6 0.4 0.2 0.0 D ev el op m en ta l r at e 0 5 0.25 Temperature (°C) 0.2 0.15 0.1 0.05 3510 20 30 (c) 15 25 y = 0.0081x – 0.05 R 2 = 0.6838 D ev el op m en ta l r at e 0.08 18 0.2 Temperature (°C) 0.18 0.16 2820 22 24 26 (b) 0.14 0.12 0.1 y = 0.0124x – 0.1384 R 2 = 0.9753 y = 0.072x – 0.32 R 2 = 0.64 Figure 2.4 Effectively linear relationships between rates of growth and development and temperature. (a) Growth of the protist Strombidinopsis multiauris. (After Montagnes et al., 2003.) (b) Egg development in the beetle Oulema duftschmidi. (After Severini et al., 2003.) (c) Egg to adult development in the mite Amblyseius californicus. (After Hart et al., 2002.) The vertical scales in (b) and (c) represent the proportion of total development achieved in 1 day at the temperature concerned. (D iff er en ce fr om V 15 )/ V 15 –0.8 –20 1.2 Temperature (°C – 15) 20–10 0 10 0.8 0.4 0 –0.4 Figure 2.5 The temperature–size rule (final size decreases with increasing temperature) illustrated in protists (65 data sets combined). The horizontal scale measures temperature as a deviation from 15°C. The vertical scale measures standardized size: the difference between the cell volume observed and the cell volume at 15°C, divided by cell volume at 15°C. The slope of the mean regression line, which must pass through the point (0,0), was −0.025 (SE, 0.004); the cell volume decreased by 2.5% for every 1°C rise in rearing temperature. (After Atkinson et al., 2003.) EIPC02 10/24/05 1:44 PM Page 34 CONDITIONS 37 ‘chilling injury’. The fruits of the banana blacken and rot after exposure to chilling temperatures and many tropical rainforest species are sensitive to chilling. The nature of the injury is obscure, although it seems to be associated with the breakdown of membrane permeability and the leakage of specific ions such as calcium (Minorsky, 1985). Temperatures below 0°C can have lethal physical and chem- ical consequences even though ice may not be formed. Water may ‘supercool’ to temperatures at least as low as −40°C, remaining in an unstable liquid form in which its physical properties change in ways that are bound to be biologically significant: its viscosity increases, its diffusion rate decreases and its degree of ionization of water decreases. In fact, ice seldom forms in an organism until the temperature has fallen several degrees below 0°C. Body fluids remain in a supercooled state until ice forms suddenly around particles that act as nuclei. The concentration of solutes in the remaining liquid phase rises as a consequence. It is very rare for ice to form within cells and it is then inevitably lethal, but the freezing of extracellular water is one of the factors that prevents ice forming within the cells themselves (Wharton, 2002), since water is withdrawn from the cell, and solutes in the cytoplasm (and vacuoles) become more concentrated. The effects of freez- ing are therefore mainly osmoregulatory: the water balance of the cells is upset and cell membranes are destabilized. The effects are essentially similar to those of drought and salinity. Organisms have at least two differ- ent metabolic strategies that allow survival through the low temperatures of winter. A ‘freeze-avoiding’ strategy uses low-molecular-weight polyhydric alcohols (polyols, such as glycerol) that depress both the freezing and the supercooling point and also ‘thermal hysteresis’ proteins that prevent ice nuclei from forming (Figure 2.8a, b). A contrasting ‘freeze-tolerant’ strategy, which also involves the formation of polyols, encour- ages the formation of extracellular ice, but protects the cell membranes from damage when water is withdrawn from the cells (Storey, 1990). The tolerances of organisms to low temperatures are not fixed but are preconditioned by the experience of tem- peratures in their recent past. This process is called acclimation when it occurs in the laboratory and acclimatization when it occurs naturally. Acclimatization may start as the weather becomes colder in the fall, stimulating the conversion of almost the entire glycogen reserve of animals into polyols (Figure 2.8c), but this can be an energetically costly affair: about 16% of the carbohydrate reserve may be consumed in the conversion of the glycogen reserves to polyols. The exposure of an individual for several days to a relatively low tem- perature can shift its whole temperature response downwards along the tem- perature scale. Similarly, exposure to a high temperature can shift the temperature response upwards. Antarctic springtails (tiny arthropods), for instance, when taken from ‘summer’ temperat- ures in the field (around 5°C in the Antarctic) and subjected to a range of acclimation temperatures, responded to temperatures in the range +2°C to −2°C (indicative of winter) by showing a marked drop in the temperature at which they froze (Figure 2.9); but at lower acclimation temperatures still (−5°C, −7°C), they showed no such drop because the temperatures were themselves too low for the physiological processes required to make the acclimation response. Acclimatization aside, individuals commonly vary in their temperature response depending on the stage of development they have reached. Probably the most extreme form of this is when an organism has a dormant stage in its life cycle. Dormant stages are typically dehydrated, metabolically slow and tolerant of extremes of temperature. 2.3.5 Genetic variation and the evolution of cold tolerance Even within species there are often differences in temperature response between populations from different locations, and these differences have frequently been found to be the result of genetic differences rather than being attributable solely to acclimatization. Powerful evidence that cold tolerance varies between geographic races of a species comes from a study of the cactus, Opuntia fragilis. Cacti are generally species of hot dry habitats, but O. fragilis extends as far north as 56°N and at one site the lowest extreme minimum temperature recorded was −49.4°C. Twenty populations were sampled from diverse localities in northern USA and Canada, and were tested for freezing tolerance and ability to acclimate to cold. Individuals from the most freeze-tolerant population (from Manitoba) tolerated −49°C in laboratory tests and acclimated by 19.9°C, whereas plants from a population in the more equable climate of Hornby Island, British Columbia, tolerated only −19°C and acclimated by only 12.1°C (Loik & Nobel, 1993). There are also striking cases where the geographic range of a crop species has been extended into colder regions by plant breeders. Programs of deliberate selection applied to corn (Zea mays) have expanded the area of the USA over which the crop can be profitably grown. From the 1920s to the 1940s, the pro- duction of corn in Iowa and Illinois increased by around 24%, whereas in the colder state of Wisconsin it increased by 54%. If deliberate selection can change the tolerance and distribu- tion of a domesticated plant we should expect natural selection to have done the same thing in nature. To test this, the plant Umbilicus rupestris, which lives in mild maritime areas of Great Britain, was deliberately grown outside its normal range (Wood- ward, 1990). A population of plants and seeds was taken from a donor population in the mild-wintered habitat of Cardiff in the west and introduced in a cooler environment at an altitude of •••• freeze-avoidance and freeze-tolerance acclimation and acclimatization EIPC02 10/24/05 1:44 PM Page 37 •• •• 38 CHAPTER 2 Te m pe ra tu re ( °C ) –40 –20 0 20 (b) DecOctSep Nov AprMarFebJan G ly ce ro l c on ce nt ra tio n (µ m ol g –1 ) 0 1000 2000 3000 (a) DecOctSep Nov AprMarFebJan G ly co ge n co nc en tr at io n (µ m ol g –1 ) 0 400 800 1200 (c) DecOctSep Nov AprMarFebJan Month Figure 2.8 (a) Changes in the glycerol concentration per gram wet mass of the freeze-avoiding larvae of the goldenrod gall moth, Epiblema scudderiana. (b) The daily temperature maxima and minima (above) and whole larvae supercooling points (below) over the same period. (c) Changes in glycogen concentration over the same period. (After Rickards et al., 1987.) EIPC02 10/24/05 1:44 PM Page 38 •• CONDITIONS 39 157 m in Sussex in the south. After 8 years, the temperature response of seeds from the donor and the introduced populations had diverged quite strikingly (Figure 2.10a), and subfreezing temperatures that kill in Cardiff (−12°C) were then tolerated by 50% of the Sussex population (Figure 2.10b). This suggests that past climatic changes, for example ice ages, will have changed the temperature tolerance of species as well as forcing their migration. •• –6 –10 –14 –22 S up er co ol in g po in t ( °C ) Exposure temperature (°C) 1 –20 5 –3 –7 –8 –12 –18 –16 –5–13 Figure 2.9 Acclimation to low temperatures. Samples of the Antarctic springtail Cryptopygus antarcticus were taken from field sites in the summer (c. 5°C) on a number of days and their supercooling point (at which they froze) was determined either immediately (
) or after a period of acclimation (
) at the temperatures shown. The supercooling points of the controls themselves varied because of temperature variations from day to day, but acclimation at temperatures in the range +2 to −2°C (indicative of winter) led to a drop in the supercooling point, whereas no such drop was observed at higher temperatures (indicative of summer) or lower temperatures (too low for a physiological acclimation response). Bars are standard errors. (After Worland & Convey, 2001.) G er m in at io n (% ) 2216 0 6 40 80 10 Temperature (°C) (a) 2 1 S ur vi va l ( % ) –14–8 0 40 80 –4 Minimum temperature (°C) (b) 2 –12 1 Figure 2.10 Changes in the behavior of populations of the plant Umbilicus rupestris, established for a period of 8 years in a cool environment in Sussex from a donor population in a mild-wintered area in South Wales (Cardiff, UK). (a) Temperature responses of seed germination: (1) responses of samples from the donor population (Cardiff ) in 1978, and (2) responses from the Sussex population in 1987. (b) The low-temperature survival of the donor population at Cardiff, 1978 (1) and of the established population in Sussex, 1987 (2). (After Woodward, 1990.) EIPC02 10/24/05 1:44 PM Page 39 •••• 42 CHAPTER 2 –2 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 Year 2000 Niño 3.4 region (threshold − 0°C) 2 1 0 –1 S ea s ur fa ce te m pe ra tu re a no m al ie s 3 (a) Figure 2.11 (a) The El Niño–Southern Oscillation (ENSO) from 1950 to 2000 as measured by sea surface temperature anomalies (differences from the mean) in the equatorial mid-Pacific. The El Niño events (> 0.4°C above the mean) are shown in dark color, and the La Niña events (> 0.4°C below the mean) are shown in pale color. (Image from http://www.cgd.ucar.edu/cas/catalog/ climind/Nino_3_3.4_indices.html.) (b) Maps of examples of El Niño (November 1997) and La Niña (February 1999) events in terms of sea height above average levels. Warmer seas are higher; for example, a sea height 15–20 cm below average equates to a temperature anomaly of approximately 2–3°C. (Image from http://topex-www.jpl.nasa.gov/science/images/el-nino-la-nina.jpg.) (For color, see Plate 2.1, between pp. 000 and 000.) (b) EIPC02 10/24/05 1:44 PM Page 42 •••• CONDITIONS 43 –4 1860 1880 1900 1920 1940 1960 1980 6 (c) Year (L n – S n) 2000 2 4 0 –2 Figure 2.11 (continued) (c) The North Atlantic Oscillation (NAO) from 1864 to 2003 as measured by the normalized sea-level pressure difference (Ln − Sn) between Lisbon, Portugal and Reykjavik, Iceland. (Image from http://www.cgd.ucar.edu/~jhurrell/ nao.stat.winter.html#winter.) (d) Typical winter conditions when the NAO index is positive or negative. Conditions that are more than usually warm, cold, dry or wet are indicated. (Image from http://www.ldeo.columbia.edu/NAO/.) (For color, see Plate 2.2, between pp. 000 and 000.) (d)(i) (d)(ii) EIPC02 10/24/05 1:44 PM Page 43 44 CHAPTER 2 isotherm (Figure 2.14a; an isotherm is a line on a map joining places that experience the same temperature – in this case a January mean of 4.5°C). However, we need to be very careful how we inter- pret such relationships: they can be extremely valuable in predicting where we might and might not find a particular species; they may suggest that some feature related to temperature is import- ant in the life of the organisms; but they do not prove that tem- perature causes the limits to a species’ distribution. The literature relevant to this and many other correlations between temperature and distribution patterns is reviewed by Hengeveld (1990), who also describes a more subtle graphical procedure. The minimum temperature of the coldest month and the maximum temperature of the hottest month are estimated for many places within and outside the range of a species. Each location is then plotted on a graph of maximum against minimum temperature, and a line is drawn that optimally discriminates between the presence and absence records (Figure 2.14b). This line is then used to define the geographic margin of the species distributions (Figure 2.14c). This may have powerful predictive value, but it still tells us nothing about the underlying forces that cause the distribution patterns. One reason why we need to be cautious about reading too much into correlations of species distributions with maps of tem- perature is that the temperatures measured for constructing isotherms for a map are only rarely those that the organisms expe- rience. In nature an organism may choose to lie in the sun or hide •••• lo g( ab un da nc e ag e 3 in 1 00 0s ) 4.5 –5 8.0 NAO index 7.5 7.0 6.5 6.0 5.5 5.0 6–4 –3 –2 –1 0 1 2 3 4 5 lo g( ab un da nc e ag e 3 in 1 00 0s ) 4.5 50 8.0 Length of 5-month-old cod (mm) 7.5 7.0 6.5 6.0 5.5 5.0 10060 70 80 90 T em pe ra tu re ( °C ) 2.5 –5 5 NAO index 4.5 4 3.5 3 6–4 –3 –2 –1 0 1 2 3 4 5 Le ng th o f 5 -m on th -o ld c od ( m m ) 50 2.5 100 Temperature (°C) 90 80 70 60 53 3.5 4 4.5 (a) (d) (b) (c) Figure 2.12 (a) The abundance of 3-year-old cod, Gadus morhua, in the Barents Sea is positively correlated with the value of the North Atlantic Oscillation (NAO) index for that year. The mechanism underlying this correlation is suggested in (b–d). (b) Annual mean temperature increases with the NAO index. (c) The length of 5-month-old cod increases with annual mean temperature. (d) The abundance of cod at age 3 increases with their length at 5 months. (After Ottersen et al., 2001.) EIPC02 10/24/05 1:44 PM Page 44 CONDITIONS 47 of only 50%. The organisms most obviously affected by humid- ity in their distribution are those ‘terrestrial’ animals that are actually, in terms of the way they control their water balance, ‘aquatic’. Amphibians, terrestrial isopods, nematodes, earthworms and molluscs are all, at least in their active stages, confined to microenvironments where the relative humidity is at or very close to 100%. The major group of animals to escape such confinement are the terrestrial arthropods, especially insects. Even here though, the evaporative loss of water often confines their activities to habitats (e.g. woodlands) or times of day (e.g. dusk) when relative humidity is relatively high. 2.5 pH of soil and water The pH of soil in terrestrial environments or of water in aquatic ones is a condition that can exert a powerful influence on the dis- tribution and abundance of organisms. The protoplasm of the root cells of most vascular plants is damaged as a direct result of toxic concentrations of H+ or OH− ions in soils below pH 3 or above pH 9, respectively. Further, indirect effects occur because soil pH influences the availability of nutrients and/or the concentration of toxins (Figure 2.17). Increased acidity (low pH) may act in three ways: (i) directly, by upsetting osmoregulation, enzyme activity or gaseous exchange across respiratory surfaces; (ii) indirectly, by increasing the con- centration of toxic heavy metals, particularly aluminum (Al3+) but also manganese (Mn2+) and iron (Fe3+), which are essential plant nutrients at higher pHs; and (iii) indirectly, by reducing the qual- ity and range of food sources available to animals (e.g. fungal growth is reduced at low pH in streams (Hildrew et al., 1984) and the aquatic flora is often absent or less diverse). Tolerance limits for pH vary amongst plant species, but only a minority are able to grow and reproduce at a pH below about 4.5. In alkaline soils, iron (Fe3+) and phosphate (PO4 3+), and certain trace elements such as manganese (Mn2+), are fixed in relatively insoluble compounds, and plants may then suffer because there is too little rather than too much of them. For example, calcifuge plants (those characteristic of acid soils) commonly show symp- toms of iron deficiency when they are transplanted to more alka- line soils. In general, however, soils and waters with a pH above 7 tend to be hospitable to many more species than those that are more acid. Chalk and limestone grasslands carry a much richer flora (and associated fauna) than acid grasslands and the situation is similar for animals inhabiting streams, ponds and lakes. Some prokaryotes, especially the Archaebacteria, can tolerate and even grow best in environments with a pH far outside the range tolerated by eukaryotes. Such environments are rare, but occur in volcanic lakes and geothermal springs where they are •••• 1.0 0 S ur vi va l r at e fu nc tio n Experiment period (days) 100 200 0.5 0 100 2000 6°C 12°C S. malma S. leucomaenis Figure 2.16 Changing temperature reverses the outcome of competition. At low temperature (6°C) on the left, the salmonid fish Salvelinus malma outsurvives cohabiting S. leucomaenis, whereas at 12°C, on the right, S. leucomaenis drives S. malma to extinction. Both species are quite capable, alone, of living at either temperature. (After Taniguchi & Nakano, 2000.) 964 5 pH 3 87 Mo Fe and Mn Cu and Zn K Ca and Mg P and B N and S mobilization Al H+ and OH– toxicity Fgiure 2.17 The toxicity of H+ and OH− to plants, and the availability to them of minerals (indicated by the widths of the bands) is influenced by soil pH. (After Larcher, 1980.) EIPC02 10/24/05 1:44 PM Page 47 48 CHAPTER 2 dominated by sulfur-oxidizing bacteria whose pH optima lie between 2 and 4 and which cannot grow at neutrality (Stolp, 1988). Thiobacillus ferroxidans occurs in the waste from industrial metal- leaching processes and tolerates pH 1; T. thiooxidans cannot only tolerate but can grow at pH 0. Towards the other end of the pH range are the alkaline environments of soda lakes with pH values of 9–11, which are inhabited by cyanobacteria such as Anabaenopsis arnoldii and Spirulina platensis; Plectonema nostocorum can grow at pH 13. 2.6 Salinity For terrestrial plants, the concentration of salts in the soil water offers osmotic resistance to water uptake. The most extreme saline conditions occur in arid zones where the predominant movement of soil water is towards the surface and cystalline salt accumu- lates. This occurs especially when crops have been grown in arid regions under irrigation; salt pans then develop and the land is lost to agriculture. The main effect of salinity is to create the same kind of osmoregulatory problems as drought and freezing and the problems are countered in much the same ways. For example, many of the higher plants that live in saline environ- ments (halophytes) accumulate electrolytes in their vacuoles, but maintain a low concentration in the cytoplasm and organelles (Robinson et al., 1983). Such plants maintain high osmotic pres- sures and so remain turgid, and are protected from the damaging action of the accumulated electrolytes by polyols and membrane protectants. Freshwater environments present a set of specialized environ- mental conditions because water tends to move into organisms from the environment and this needs to be resisted. In marine habitats, the majority of organisms are isotonic to their environ- ment so that there is no net flow of water, but there are many that are hypotonic so that water flows out from the organism to the environment, putting them in a similar position to terrestrial organisms. Thus, for many aquatic organisms the regulation of body fluid concentration is a vital and sometimes an energetically expensive process. The salinity of an aquatic environment can have an important influence on distribution and abundance, especially in places like estuaries where there is a particularly sharp gradi- ent between truly marine and freshwater habitats. The freshwater shrimps Palaemonetes pugio and P. vulgaris, for example, co-occur in estuaries on the eastern coat of the USA at a wide range of salinities, but the former seems to be more tolerant of lower salinities than the latter, occupying some habitats from which the latter is absent. Figure 2.18 shows the mechanism likely to be underlying this (Rowe, 2002). Over the low salinity range (though not at the effectively lethal lowest salin- ity) metabolic expenditure was significantly lower in P. pugio. P. vulgaris requires far more energy simply to maintain itself, putting it at a severe disadvantage in competition with P. pugio even when it is able to sustain such expenditure. 2.6.1 Conditions at the boundary between the sea and land Salinity has important effects on the distribution of organisms in intertidal areas but it does so through interactions with other conditions – notably exposure to the air and the nature of the substrate. •••• S ta nd ar d m et ab ol ic e xp en di tu re ( J da y– 1 ) 33 32 31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 Salinity (ppt) 0 1 2 3 4 5 6 7 353025201510 Overall mean, P. vulgaris (24.85) Overall mean, P. pugio (22.91) P. pugio P. vulgaris Figure 2.18 Standard metabolic expenditure (estimated through minimum oxygen consumption) in two species of shrimp, Palaemonetes pugio and P. vulgaris, at a range of salinities. There was significant mortality of both species over the experimental period at 0.5 ppt (parts per thousand), especially in P. vulgaris (75% compared with 25%). (After Rowe, 2002.) EIPC02 10/24/05 1:44 PM Page 48 CONDITIONS 49 Algae of all types have found suitable habitats permanently immersed in the sea, but permanently submerged higher plants are almost completely absent. This is a striking contrast with submerged freshwater habitats where a variety of flowering plants have a conspicuous role. The main reason seems to be that higher plants require a substrate in which their roots can find anchorage. Large marine algae, which are continuously sub- merged except at extremely low tides, largely take their place in marine communities. These do not have roots but attach themselves to rocks by specialized ‘holdfasts’. They are excluded from regions where the substrates are soft and holdfasts cannot ‘hold fast’. It is in such regions that the few truly marine flower- ing plants, for example sea grasses such as Zostera and Posidonia, form submerged communities that support complex animal communities. Most species of higher plants that root in seawater have leaves and shoots that are exposed to the atmosphere for a large part of the tidal cycle, such as mangroves, species of the grass genus Spartina and extreme halo- phytes such as species of Salicornia that have aerial shoots but whose roots are exposed to the full salinity of seawater. Where there is a stable substrate in which plants can root, communities of flowering plants may extend right through the intertidal zone in a continuum extending from those continuously immersed in full-strength seawater (like the sea grasses) through to totally non- saline conditions. Salt marshes, in particular, encompass a range of salt concentrations running from full-strength seawater down to totally nonsaline conditions. Higher plants are absent from intertidal rocky sea shores except where pockets of soft substrate may have formed in crevices. Instead, such habitats are dominated by the algae, which give way to lichens at and above the high tide level where the exposure to desiccation is highest. The plants and animals that live on rocky sea shores are influenced by environmental condi- tions in a very profound and often particularly obvious way by the extent to which they tolerate exposure to the aerial environ- ment and the forces of waves and storms. This expresses itself in the zonation of the organisms, with different species at different heights up the shore (Figure 2.19). The extent of the intertidal zone depends on the height of tides and the slope of the shore. Away from the shore, the tidal rise and fall are rarely greater than 1 m, but closer to shore, the shape of the land mass can funnel the ebb and flow of the water to produce extraordinary spring tidal ranges of, for example, nearly 20 m in the Bay of Fundy (between Nova Scotia and New Brunswick, Canada). In contrast, the shores of the Mediterranean Sea •••• Figure 2.19 A general zonation scheme for the seashore determined by relative lengths of exposure to the air and to the action of waves. (After Raffaelli & Hawkins, 1996.) Land Sea Supra littora l zone Up per lim it o f la min atio n s eaw eed s Upp er li mit of b arna cles Uppe r lim it of periw inkle sna ils Supralittoral fringe Midlittoral zone Infralittoral zone Infralittoral fringe Littoral zone algae and higher plants zonation EIPC02 10/24/05 1:44 PM Page 49 •• 52 CHAPTER 2 recorded in the succession of diatom species accumulated in lake sediments (Flower et al., 1994). Figure 2.22, for example, shows how diatom species composition has changed in Lough Maam, Ireland – far from major industrial sites. The percentage of vari- ous diatom species at different depths reflects the flora present at various times in the past (four species are illustrated). The age of layers of sediment can be determined by the radioactive decay of lead-210 (and other elements). We know the pH tolerance of the diatom species from their present distribution and this can be used to reconstruct what the pH of the lake has been in the past. Note how the waters acidified since about 1900. The diatoms Fragilaria virescens and Brachysira vitrea have declined markedly dur- ing this period while the acid-tolerant Cymbella perpusilla and Frustulia rhomboides increased after 1900. 2.9 Global change In Chapter 1 we discussed some of the ways in which global environments have changed over the long timescales involved in continental drift and the shorter timescales of the repeated ice ages. Over these timescales some organisms have failed to accommodate to the changes and have become extinct, others have migrated so that they continue to experience the same conditions but in a different place, and it is probable that others have changed their nature (evolved) and tolerated some of the changes. We now turn to consider global changes that are occur- ring in our own lifetimes – consequences of our own activities – and that are predicted, in most scenarios, to bring about profound changes in the ecology of the planet. 2.9.1 Industrial gases and the greenhouse effect A major element of the Industrial Revolution was the switch from the use of sustainable fuels to the use of coal (and later, oil) as a source of power. Between the middle of the 19th and the middle of the 20th century the burning of fossil fuels, together with extensive deforestation, added about 9 × 1010 tonnes of carbon dioxide (CO2) to the atmosphere and even more has been added since. The concentration of CO2 in the atmosphere before the Industrial Revolution (measured in gas trapped in ice cores) was about 280 ppm, a fairly typical interglacial ‘peak’ (Figure 2.23), but this had risen to around 370 ppm by around the turn of the millennium and is still rising (see Figure 18.22). Solar radiation incident on the earth’s atmosphere is in part reflected, in part absorbed, and part is transmitted through to the earth’s surface, which absorbs and is warmed by it. Some of this absorbed energy is radiated back to the atmosphere where atmo- spheric gases, mainly water vapor and CO2 absorb about 70% of it. It is this trapped reradiated energy that heats the atmosphere in what is called the ‘greenhouse effect’. The greenhouse effect was of course part of the normal environment before the Industrial Revolution and carried responsibility for some of the environmental warmth before industrial activity started to enhance it. At that time, atmospheric water vapor was respons- ible for the greater portion of the greenhouse effect. •• 25 10 10 25 100 250 100 25 10 500 1000 100 250 1000 2000 3000 4000 1000 100250 500 500 2000 10 00 500 Figure 2.21 An example of long-distance environmental pollution. The distribution in Great Britain of fallout of radioactive caesium (Bq m−2) from the Chernobyl nuclear accident in the Soviet Union in 1986. The map shows the persistence of the pollutant on acid upland soils where it is recycled through soils, plants and animals. Sheep in the upland areas contained more caesium-137 (137Cs) in 1987 and 1988 (after recycling) than in 1986. 137Cs has a half-life of 30 years! On typical lowland soils it is more quickly immobilized and does not persist in the food chains. (After NERC, 1990.) EIPC02 10/24/05 1:44 PM Page 52 •• CONDITIONS 53 In addition to the enhancement of greenhouse effects by increased CO2, other trace gases have increased markedly in the atmosphere, particularly methane (CH4) (Figure 2.24a; and compare this with the his- torical record in Figure 2.23), nitrous oxide (N2O) and the chlorofluorocarbons (CFCs, e.g. trichlorofluoromethane (CCl3F) and dichlorodifluoromethane (CCl2F2)). Together, these and other gases contribute almost as much to enhancing the green- house effect as does the rise in CO2 (Figure 2.24b). The increase in CH4 is not all explained but probably has a microbial origin in intensive agriculture on anaerobic soils (especially increased rice production) and in the digestive process of ruminants (a cow pro- duces approximately 40 litres of CH4 each day); around 70% of its production is anthropogenic (Khalil, 1999). The effect of the CFCs from refrigerants, aerosol propellants and so on is poten- tially great, but international agreements at least appear to have halted further rises in their concentrations (Khalil, 1999). It should be possible to draw up a balance sheet that shows how the CO2 produced by human activities translates into the changes in concentration in the atmosphere. Human activities •• Percent 0 3010 B ra ch ys ira v itr ea 20 Date A.D. 1988 1969 1940 1903 5.2 5.4 5.6 5.8 6.0 pH 0 10 20 30 40 0 3010 F ra gi la ria v ire sc en s 200 10 F ru st ul ia r ho m bo id es S ed im en t d ep th ( cm ) 0 40 5 10 15 30 25 20 0 20 35 10 C ym be lla p er pu si lla Figure 2.22 The history of the diatom flora of an Irish lake (Lough Maam, County Donegal) can be traced by taking cores from the sediment at the bottom of the lake. The percentage of various diatom species at different depths reflects the flora present at various times in the past (four species are illustrated). The age of the layers of sediment can be determined by the radioactive decay of lead-210 (and other elements). We know the pH tolerance of the diatom species from their present distribution and this can be used to reconstruct what the pH of the lake has been in the past. Note how the waters have been acidified since about 1900. The diatoms Fragilaria virescens and Brachysira vitrea have declined markedly during this period, while the acid-tolerant Cymbella perpusilla and Frustulia rhomboides have increased. (After Flower et al., 1994.) CO2 – but not only CO2 C O 2 (p pm ) 0100,000300,000400,000 200 240 280 200,000 Age BP (years) C H 4 (p pb ) 400 600 700 500 Figure 2.23 Concentrations of CO2 and methane (CH4) in gas trapped in ice cores from Vostok, Antarctica deposited over the past 420,000 years. Estimated temperatures are very strongly correlated with these. Thus, transitions between glacial and warm epochs occurred around 335,000, 245,000, 135,000 and 18,000 years ago. BP, before present; ppb, parts per billion; ppm, parts per million. (After Petit et al., 1999; Stauffer, 2000.) EIPC02 10/24/05 1:44 PM Page 53 54 CHAPTER 2 release 5.1–7.5 × 109 metric tons of carbon to the atmosphere each year. But the increase in atmospheric CO2 (2.9 × 109 metric tons) accounts for only 60% of this, a percentage that has remained remarkably constant for 40 years (Hansen et al., 1999). The oceans absorb CO2 from the atmosphere, and it is estimated that they may absorb 1.8–2.5 × 109 metric tons of the carbon released by human activities. Recent analyses also indicate that terrestrial vegetation has been ‘fertilized’ by the increased atmospheric CO2, so that a considerable amount of extra carbon has been locked up in vegetation biomass (Kicklighter et al., 1999). This softening of the blow by the oceans and terrestrial vegetation notwith- standing, however, atmospheric CO2 and the greenhouse effect are increasing. We return to the question of global carbon budgets in Section 18.4.6. 2.9.2 Global warming We started this chapter discussing temperature, moved through a number of other environmental conditions to pollutants, and now return to temperature because of the effects of those pollu- tants on global temperatures. It appears that the present air temperature at the land surface is 0.6 ± 0.2°C warmer than in preindustrial times (Figure 2.25), and temperatures are predicted to continue to rise by a further 1.4–5.8°C by 2100 (IPCC, 2001). Such changes will probably lead to a melting of the ice caps, a consequent rising of sea level and large changes in the pattern of global climates and the distribution of species. Predictions of the extent of global warming resulting from the enhanced green- house effect come from two sources: (i) predictions based on sophisticated computer models (‘general circulation models’) that simulate the world’s climate; and (ii) trends detected in mea- sured data sets, including the width of tree rings, sea-level records and measures of the rate of retreat of glaciers. Not surprisingly, different global circulation models differ in their pre- dictions of the rise in global tempera- ture that will result from predicted increases in CO2. However, most model predictions vary only from 2.3 to 5.2°C (most of the variation is accounted for by the way in which the effects of cloud cover are modeled), and a projected rise of 3–4°C in the next 100 years seems a reasonable value from which to make projections of ecological effects (Figure 2.26). But temperature regimes are, of course, only part of the set of conditions that determine which organisms live where. Unfortunately, we can place much less faith in computer projec- tions of rainfall and evaporation because it is very hard to build good models of cloud behavior into a general model of climate. If we consider only temperature as a relevant variable, we would project a 3°C rise in temperature giving London (UK) the climate of Lisbon (Portugal) (with an appropriate vegetation of olives, vines, Bougainvillea and semiarid scrub). But with more reliable rain it would be nearly subtropical, and with a little less it might qualify for the status of an arid zone! •••• C on ce nt ra te d C H 4 (p pb ) 200019601920 800 1900 1400 1600 1800 1940 Year (a) 1000 1200 1980 C al cu la te d te m pe ra tu re c ha ng e (° C ) 0.0 CO2 0.3 0.4 0.5 Trace gas (b) 0.1 0.2 CFCsCH4 N2O Figure 2.24 (a) Concentration of methane (CH4) in the atmosphere through the 20th century. (b) Estimates of global warming over the period 1850–1990 caused by CO2 and other major greenhouse gases. (After Khalil, 1999.) a 3–4°C rise in the next 100 years EIPC02 10/24/05 1:44 PM Page 54 CONDITIONS 57 modern concept – Hutchinson’s n-dimensional hypervolume – also distinguishes fundamental and realized niches. Temperature is discussed in detail as a typical, and perhaps the most important, condition. Individuals respond to temperature with impaired function and ultimately death at upper and lower extremes, with a functional range between the extremes, within which there is an optimum, although these responses may be subject to evolutionary adaptation and to more immediate acclimatization. The rates of biological enzymatic processes often increase exponentially with temperature (often Q10 ≈ 2), but for rates of growth and development there are often only slight deviations from linearity: the basis for the day-degree concept. Because development usually increases more rapidly with temperature than does growth, final size tends to decrease with rearing temperature. Attempts to uncover universal rules of temperature dependence remain a matter of controversy. We explain the differences between endotherms and ecto- therms but also the similarities between them, ultimately, in their responses to a range of temperatures. We examine variations in temperature on and within the sur- face of the earth with a variety of causes: latitudinal, altitudinal, continental, seasonal, diurnal and microclimatic effects, and, in soil and water, the effects of depth. Increasingly, the importance of medium-term temporal patterns have become apparent. Notable amongst these are the El Niño–Southern Oscillation (ENSO) and the North Atlantic Oscillation (NAO). There are very many examples of plant and animal distribu- tions that are strikingly correlated with some aspect of environ- mental temperature but these do not prove that temperature directly causes the limits to a species’ distribution. The tem- peratures measured are only rarely those that the organisms experience. For many species, distributions are accounted for not so much by average temperatures as by occasional extremes; and the effects of temperature may be determined largely by the responses of other community members or by interactions with other conditions. A range of other environmental conditions are also discussed: the pH of soil and water, salinity, conditions at the boundary between sea and land, and the physical forces of winds, waves and currents. Hazards, disasters and catastrophes are distinguished. A number of environmental conditions are becoming increas- ingly important due to the accumulation of toxic by-products of human activities. A striking example is the creation of ‘acid rain’. Another is the effect of industrial gases on the greenhouse effect and consequent effects on global warming. A projected rise of 3–4°C in the next 100 years seems a reasonable value from which to make projections of ecological effects, though global warming is not evenly distributed over the surface of the earth. This rate is 50–100 times faster than postglacial warming. We must expect latitudinal and altitudinal changes to species’ distributions and widespread extinctions of floras and faunas. •••• EIPC02 10/24/05 1:44 PM Page 57 •• 3.1 Introduction According to Tilman (1982), all things consumed by an organism are resources for it. But consumed does not simply mean ‘eaten’. Bees and squirrels do not eat holes, but a hole that is occupied is no longer available to another bee or squirrel, just as an atom of nitrogen, a sip of nectar or a mouthful of acorn are no longer available to other consumers. Similarly, females that have already mated may be unavailable to other mates. All these things have been consumed in the sense that the stock or supply has been reduced. Thus, resources are entities required by an organ- ism, the quantities of which can be reduced by the activity of the organism. Green plants photosynthesize and obtain both energy and matter for growth and reproduction from inorganic materials. Their resources are solar radiation, carbon dioxide (CO2), water and mineral nutrients. ‘Chemosynthetic’ organisms, such as many of the Archaebacteria, obtain energy by oxidizing methane, ammonium ions, hydrogen sulfide or ferrous iron; they live in environments such as hot springs and deep sea vents and use resources that were much more abundant during early phases of life on earth. All other organisms use as their food resource the bodies of other organisms. In each case, what has been consumed is no longer available to another consumer. The rabbit eaten by an eagle is no longer available to another eagle. The quantum of solar radiation absorbed and photosynthesized by a leaf is no longer available to another leaf. This has an important consequence: organ- isms may compete with each other to capture a share of a limited resource – a topic that will occupy us in Chapter 5. A large part of ecology is about the assembly of inorganic resources by green plants and the reassembly of these packages at each successive stage in a web of consumer–resource inter- actions. In this chapter we start with the resources of plants and focus especially on those most important in photosynthesis: radiation and CO2. Together, plant resources fuel the growth of individual plants, which, collectively, determine the primary productivity of whole areas of land (or volumes of water): the rate, per unit area, at which plants produce biomass. Patterns of prim- ary productivity are examined in Chapter 17. Relatively little space in this chapter is given to food as a resource for animals, simply because a series of later chapters (9–12) is devoted to the ecology of predators, grazers, parasites and saprotrophs (the consumers and decomposers of dead organisms). This chapter then closes where the previous chapter began: with the ecological niche, adding resource dimensions to the condition dimensions we have met already. 3.2 Radiation Solar radiation is the only source of energy that can be used in metabolic activities by green plants. It comes to the plant as a flux of radiation from the sun, either directly having been diffused to a greater or lesser extent by the atmosphere, or after being reflected or transmitted by other objects. The direct fraction is highest at low latitudes (Figure 3.1). Moreover, for much of the year in temperate climates, and for the whole of the year in arid climates, the leaf canopy in terrestrial communities does not cover the land surface, so that most of the incident radiation falls on bare branches or on bare ground. When a plant intercepts radiant energy it may be reflected (with its wavelength unchanged), transmitted (after some wavebands have been filtered out) or absorbed. Part of the fraction that is absorbed may raise the plant’s temperature and be reradiated at much longer wavelengths; in terrestrial plants, part may contribute latent heat of evaporation of water and so power the transpiration what are resources? organisms may compete for resources the fate of radiation Chapter 3 Resources EIPC03 10/24/05 1:47 PM Page 58 RESOURCES 59 stream. A small part may reach the chloroplasts and drive the process of photosynthesis (Figure 3.2). Radiant energy is converted during photosynthesis into energy-rich chem- ical compounds of carbon, which will subsequently be broken down in re- spiration (either by the plant itself or by organisms that consume it). But unless the radiation is cap- tured and chemically fixed at the instant it falls on the leaf, it is irretrievably lost for photosynthesis. Radiant energy that has been fixed in photosynthesis passes just once through the world. This is in complete contrast to an atom of nitrogen or carbon or a molecule of water that may cycle repeatedly through endless generations of organisms. Solar radiation is a resource con- tinuum: a spectrum of different wave- lengths. But the photosynthetic apparatus is able to gain access to energy in only a restricted band of this spectrum. All green plants depend on chlorophyll and other pigments for the photosynthetic fixation of carbon, and these pigments fix radiation in a waveband between roughly 400 and 700 nm. This is the band of ‘photosynthetically active radiation’ (PAR). It corresponds broadly with the range of the spectrum visible to the human eye that we call ‘light’. About 56% of the radiation incident on the earth’s surface lies outside the PAR range and is thus unavailable as a resource for green plants. In other organisms there are pigments, for example bacterio- chlorophyll in bacteria, that operate in photosynthesis outside the PAR range of green plants. 3.2.1 Variations in the intensity and quality of radiation A major reason why plants seldom achieve their intrinsic photosynthetic capacity is that the intensity of radiation varies continually (Figure 3.3). Plant morphology and physiology that are optimal for photosynthesis at one intensity of radiation will usually be inappropriate at another. In terrestrial habitats, leaves live in a radiation regime that varies throughout the day and the year, and they live in an environment of other leaves that modifies the quantity and quality of radiation received. As with all resources, the supply of radiation can vary both systematically (diurnal, annual) and unsystematically. Moreover, it is not the case simply that the inten- sity of radiation is a greater or lesser proportion of a maximum value at which photosynthesis would be most productive. At high intensities, photoinhibition of photosynthesis may occur (Long et al., 1994), such that the rate of fixation of carbon decreases with increasing radiation intensity. High intensities of radiation may also lead to dangerous overheating of plants. Radiation is an essential resource for plants, but they can have too much as well as too little. Annual and diurnal rhythms are systematic variations in solar radiation (Figure 3.3a, b). The green plant expe- riences periods of famine and glut in its radiation resource every 24 h (except near the poles) and seasons of famine and glut every year (except in the tropics). In aquatic habitats, an additional •• radiant energy must be captured or is lost forever photosynthetically active radiation 2.1 2.1 1.68 1.68 1.68 1.68 1.26 0.84 2.1 2.1 1.68 1.68 1.68 1.68 2.1 1.68 1.26 0.84 0.84 1.26 2.1 1.68 1.26 1.68 2.12.1 2.1 1.68 0.84 Figure 3.1 Global map of the solar radiation absorbed annually in the earth– atmosphere system: from data obtained with a radiometer on the Nimbus 3 meteorological satellite. The units are J cm−2 min−1. (After Raushke et al., 1973.) photoinhibition at high intensities systematic variations in supply EIPC03 10/24/05 1:47 PM Page 59 62 CHAPTER 3 rhythm in the intensity of radiation – they are shed when they are least useful. In consequence, an evergreen leaf of an under- story species may experience a further systematic change, because the seasonal cycle of leaf production of overstory species deter- mines what radiation remains to penetrate to the understory. The daily movement of leaves in many species also reflects the changing intensity and direction of incident radiation. Less systematic variations in the radiation environment of a leaf are caused by the nature and position of neighboring leaves. Each canopy, each plant and each leaf, by intercepting radiation, creates a resource- depletion zone (RDZ) – a moving band of shadow over other leaves of the same plant, or of others. Deep in a canopy, shadows become less well defined because much of the radiation loses its original direction by diffusion and reflection. Submerged vegetation in aquatic habitats is likely to have a much less sys- tematic shading effect, simply because it is moved around by the flow of the water in which it lives, though vegeta- tion floating on the surface, especially of ponds or lake, inevitably has a profound and largely unvary- ing effect on the radiation regime beneath it. Phytoplankton cells nearer the surface, too, shade the cells beneath them, such that the reduction of intensity with depth is greater, the greater the phytoplankton density. Figure 3.4, for example, shows the decline in light penetration, measured at a set depth in a laboratory system, as a population of the unicellular green alga, Chlorella vulgaris, built up over a 12-day period (Huisman, 1999). The composition of radiation that has passed through leaves in a canopy, or through a body of water, is also altered. It may be less useful photo- synthetically because the PAR component has been reduced – though such reductions may also, of course, prevent photo- inhibition and overheating. Figure 3.5 shows an example for the variation with depth in a freshwater habitat. The major differences amongst ter- restrial species in their reaction to sys- tematic variations in the intensity of radiation are those that have evolved between ‘sun species’ and ‘shade species’. In general, plant species that are characteristic of shaded habitats use radiation at low intensities more efficiently than sun species, but the reverse is true at high intensities (Figure 3.6). Part of the difference between them lies in the physiology of the leaves, but the mor- phology of the plants also influences the efficiency with which radiation is captured. The leaves of sun plants are commonly exposed at acute angles to the midday sun (Poulson & DeLucia, 1993). This spreads an incident beam of radiation over a larger leaf area, and effectively reduces its intensity. An intensity of radiation that is superoptimal for photosynthesis when it strikes a leaf at 90° may therefore be optimal for a leaf inclined at an acute angle. The leaves of sun plants are often superimposed into •••• shade: a resource- depletion zone attenuation with depth, and plankton density, in aquatic habitats variations in quality as well as quantity sun and shade species 20 60 40 20 0 15 10 5 0 Time (days) 24201612840 P op ul at io n de ns ity ( ce lls m l– 1 ) × 1 0– 6 Li gh t p en et ra tio n (µ m ol p ho to ns m –2 s –1 ) Figure 3.4 As population density (
) of the unicellular green alga, Chlorella vulgaris, increased in laboratory culture, this increased density reduced the penetration of light (7; its intensity at a set depth). Bars are standard deviations; they are omitted when they are smaller than the symbols. (After Huisman, 1999.) Q ua nt um ir ra di an ce (1 01 5 qu an ta m –2 s –1 n m –1 ) 5000 4000 3000 2000 1000 0 Wavelength (nm) 750700650600550500450400 0 m 5 m (×25) 3 m Figure 3.5 Changing spectral distribution of radiation with depth in Lake Burley Griffin, Australia. Note that photosynthetically active radiation lies broadly within the range 400–700 nm. (After Kirk, 1994.) EIPC03 10/24/05 1:47 PM Page 62 RESOURCES 63 a multilayered canopy. In bright sunshine even the shaded leaves in lower layers may have positive rates of net photosynthesis. Shade plants commonly have leaves held near to the horizontal and in a single-layered canopy. In contrast to these ‘strategic’ dif- ferences, it may also happen that as a plant grows, its leaves develop differently as a ‘tactical’ response to the radiation environment in which it developed. This often leads to the formation of ‘sun leaves’ and ‘shade leaves’ within the canopy of a single plant. Sun leaves are typically smaller, thicker, have more cells per unit area, denser veins, more densely packed chloroplasts and a greater dry weight per unit area of leaf. These tactical maneuvers, then, tend to occur not at the level of the whole plant, but at the level of the individual leaf or even its parts. Nevertheless, they take time. To form sun or shade leaves as a tactical response, the plant, its bud or the developing leaf must sense the leaf ’s environment and respond by growing a leaf with an appropriate structure. For exam- ple, it is impossible for the plant to change its form fast enough to track the changes in intensity of radiation between a cloudy and a clear day. It can, however, change its rate of photosyn- thesis extremely rapidly, reacting even to the passing of a fleck of sunlight. The rate at which a leaf photosynthesizes also depends on the demands that are made on it by other vigorously growing parts. Photosynthesis may be reduced, even though conditions are otherwise ideal, if there is no demanding call on its products. In aquatic habitats, much of the variation between species is accounted for by differences in photosynthetic pigments, which contribute significantly to the precise wave- lengths of radiation that can be utilized (Kirk, 1994). Of the three types of pigment – chlorophylls, carotenoids and biliproteins – all photosynthetic plants contain the first two, but many algae also contain biliproteins; and within the chlorophylls, all higher plants have chlorophyll a and b, but many algae have only chlorophyll a and some have chlorophyll a and c. Examples of the absorp- tion spectra of a number of pigments, the related contrasting absorption spectra of a number of groups of aquatic plants, and the related distributional differences (with depth) between a number of groups of aquatic plants are illustrated in Figure 3.7. A detailed assessment of the evidence for direct links between pigments, performance and distribution is given by Kirk (1994). 3.2.2 Net photosynthesis The rate of photosynthesis is a gross measure of the rate at which a plant captures radiant energy and fixes it in organic carbon compounds. However, it is often more important to consider, and very much easier to measure, the net gain. Net photosynthesis is the increase (or decrease) in dry matter that results from the difference between gross photosynthesis and the losses due to respiration and the death of plant parts (Figure 3.8). Net photosynthesis is negative in darkness, when respiration exceeds photosynthesis, and increases with the intensity of PAR. The compensation point is the intensity of PAR at which the gain from gross photosyn- thesis exactly balances the respiratory and other losses. The leaves of shade species tend to respire at lower rates than those of sun species. Thus, when both are growing in the shade the net photo- synthesis of shade species is greater than that of sun species. There is nearly a 100-fold variation in the photosynthetic capacity of leaves (Mooney & Gulmon, 1979). This is the rate of photosynthesis when incident radiation is saturating, temperature is optimal, relative humidity is high, and CO2 and oxygen concentrations are normal. When the leaves of different species are compared under these ideal conditions, the ones with the highest photosynthetic capacity are generally those from environments where nutrients, water and radiation are seldom limiting (at least during the growing season). These include many agricultural crops and their weeds. Species from resource-poor environments (e.g. shade plants, desert perennials, heathland species) usually have low photosynthetic capacity – even when abundant resources are provided. Such pat- terns can be understood by noting that photosynthetic capacity, like all capacity, must be ‘built’; and the investment in building •••• C O 2 up ta ke ( m g C O 2 dm –2 h –1 ) 0 0 30 40 50 Radiation intensity (100 J m–2 s–1) 987654321 20 10 C4 C3 Shade herbs Shade mosses, planktonic algae Beech Sun herbs Wheat Corn Sorghum 10 Figure 3.6 The response of photosynthesis to light intensity in various plants at optimal temperatures and with a natural supply of CO2. Note that corn and sorghum are C4 plants and the remainder are C3 (the terms are explained in Sections 3.3.1 and 3.3.2). (After Larcher, 1980, and other sources.) sun and shade leaves the compensation point photosynthetic capacity pigment variation in aquatic species EIPC03 10/24/05 1:47 PM Page 63 •••• 64 CHAPTER 3 A bs or ba nc e Macrophyte 750 0.0 1.0 Wavelength (nm) 600500400 0.8 0.6 0.4 0.2 700650550450 (e) A bs or ba nc e (d) 700 0.0 0.9 300 Wavelength (nm) 600500400 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 R-phycocyanin (b) A bs or ba nc e 700550450 0.00 0.75 1.00 450 Wavelength (nm) 650600 0.25 449 500 0.50 628 Chlorophyll c2 A bs or ba nc e β-carotene 550 0.0 2.0 2.5 400 Wavelength (nm) 500 1.0 450 1.5 0.5 (c) A bs or ba nc e Green algae(f) (a rb itr ar y un its ) 750 Wavelength (nm) 600500400 700650550450 (a) A bs or ba nc e 700550450 0.0 1.0 1.5 500 650600 0.5 Chlorophyll b Chlorophyll a Chlorophyll a and b Wavelength (nm) Figure 3.7 (a) Absorption spectra of chlorophylls a and b. (b) Absorption spectrum of chlorophyll c2. (c) Absorption spectrum of β-carotene. (d) Absorption spectrum of the biliprotein, R-phycocyanin. (e) Absorption spectrum of a piece of leaf of the freshwater macrophyte, Vallisneria spiralis, from Lake Ginnindera, Australia. (f ) Absorption spectrum of the planktonic alga Chlorella pyrenoidos (green). EIPC03 10/24/05 1:47 PM Page 64 RESOURCES 67 Overall, therefore, despite receiving only one-seventh of the PFD of sun plants, shade plants reduced the differential in the amount absorbed to one-quarter, and reduced the differential in their daily rate of carbon gain to only a half. Shade plants successfully counterbalanced their reduced photosynthetic capa- city at the leaf level with enhanced light-harvesting ability at the whole plant level. The sun plants can be seen as striking a compromise between maximizing whole plant photosynthesis on the one hand while avoiding photoinhibition and overheating of individual leaves on the other. 3.2.4 Photosynthesis or water conservation? Strategic and tactical solutions In fact, in terrestrial habitats especially, it is not sensible to consider radiation as a resource independently of water. Intercepted radiation does not result in photosynthesis unless there is CO2 available, and the prime route of entry of CO2 is through open stomata. But if the stomata are open to the air, water will evaporate through them. If water is lost faster than it can be gained, the leaf (and the plant) will sooner or later wilt and eventually die. But in most terres- trial communities, water is, at least sometimes, in short supply. Should a plant conserve water at the expense of present photo- synthesis, or maximize photosynthesis at the risk of running out of water? Once again, we meet the problem of whether the optimal solution involves a strict strategy or the ability to make tactical responses. There are good examples of both solutions and also compromises. Perhaps the most obvious strategy that plants may adopt is to have a short life and high photosynthetic activity during periods when water is abundant, but remain dormant as seeds during the rest of the year, neither photosynthesizing nor transpiring (e.g. many desert annuals, annual weeds and most annual crop plants). •••• Table 3.1 (a) Observed differences in the shoots and leaves of sun and shade plants of the shrub Heteromeles arbutifolia. Standard deviations are given in parentheses; the significance of differences are given following analysis of variance. (b) Consequent whole plant properties of sun and shade plants. (After Valladares & Pearcy, 1998.) (a) Sun Shade P Internode distance (cm) 1.08 (0.06) 1.65 (0.02) < 0.05 Leaf angle (degrees) 71.3 (16.3) 5.3 (4.3) < 0.01 Leaf surface area (cm2) 10.1 (0.3) 21.4 (0.8) < 0.01 Leaf blade thickness (mm) 462.5 (10.9) 292.4 (9.5) < 0.01 Photosynthetic capacity, area basis (mmol CO2 m −2 s−1) 14.1 (2.0) 9.0 (1.7) < 0.01 Photosynthetic capacity, mass basis (mmol CO2 kg −1 s−1) 60.8 (10.1) 58.1 (11.2) NS Chlorophyll content, area basis (mg m−2) 280.5 (15.3) 226.7 (14.0) < 0.01 Chlorophyll content, mass basis (mg g−1) 1.23 (0.04) 1.49 (0.03) < 0.05 Leaf nitrogen content, area basis (g m−2) 1.97 (0.25) 1.71 (0.21) < 0.05 Leaf nitrogen content, mass basis (% dry weight) 0.91 (0.31) 0.96 (0.30) NS (b) Sun plants Shade plants Summer Winter Summer Winter EP 0.55 a 0.80b 0.88b 0.54a ED 0.33 a 0.38a, b 0.41b 0.43b Fraction self-shaded 0.22a 0.42b 0.47b 0.11a EA, direct PFD 0.28 a 0.44b 0.55c 0.53c LARc (cm 2 g−1) 7.1a 11.7b 20.5c 19.7c EP, projection efficiency; ED, display efficiency; EA, absorption efficiency; LARe, effective leaf area ratio; NS, not significant. Letter codes indicate groups that differed significantly in analyses of variance (P < 0.05). stomatal opening short active interludes in a dormant life EIPC03 10/24/05 1:47 PM Page 67 68 CHAPTER 3 Second, plants with long lives may produce leaves during periods when water is abundant and shed them during droughts (e.g. many species of Acacia). Some shrubs of the Israeli desert (e.g. Teucrium polium) bear finely divided, thin- cuticled leaves during the season when soil water is freely avail- able. These are then replaced by undivided, small, thick-cuticled leaves in more drought-prone seasons, which in turn fall and may leave only green spines or thorns (Orshan, 1963): a sequential polymorphism through the season, with each leaf morph being replaced in turn by a less photosynthetically active but more water- tight structure. Next, leaves may be produced that are long lived, transpire only slowly and tolerate a water deficit, but which are unable to photosynthesize rapidly even when water is abundant (e.g. evergreen desert shrubs). Structural features such as hairs, sunken stomata and the restriction of stomata to specialized areas on the lower surface of a leaf slow down water loss. But these same morphological features reduce the rate of entry of CO2. Waxy and hairy leaf surfaces may, however, reflect a greater proportion of radiation that is not in the PAR range and so keep the leaf temperature down and reduce water loss. Finally, some groups of plants have evolved particular physiologies: C4 and crassulacean acid metabolism (CAM). We consider these in more detail in Sections 3.3.1–3.3.3. Here, we simply note that plants with ‘nor- mal’ (i.e. C3) photosynthesis are wasteful of water compared with plants that possess the modified C4 and CAM physiologies. The water-use efficiency of C4 plants (the amount of carbon fixed per unit of water transpired) may be double that of C3 plants. The viability of alternative strat- egies to solve a common problem is nicely illustrated by the trees of seasonally dry tropical forests and woodlands (Eamus, 1999). These communities are found naturally in Africa, the Americas, Australia and India, and as a result of human interference elsewhere in Asia. But whereas, for example, the savannas of Africa and India are dominated by deciduous species, and the Llanos of South America are dominated by evergreens, the savannas of Australia are occu- pied by roughly equal numbers of species from four groups (Figure 3.10a): evergreens (a full canopy all year), deciduous species (losing all leaves for at least 1 and usually 2–4 months each year), semideciduous species (losing around 50% or more of their leaves each year) and brevideciduous species (losing only about 20% of their leaves). At the ends of this continuum, the decidu- ous species avoid drought in the dry season (April–November in Australia) as a result of their vastly reduced rates of transpir- ation (Figure 3.10b), but the evergreens maintain a positive carbon balance throughout the year (Figure 3.10c), whereas the deciduous species make no net photosynthate at all for around 3 months. The major tactical control of the rates of both photosynthe- sis and water loss is through changes in stomatal ‘conductance’ that may occur rapidly during the course of a day and allow a very rapid response to immediate water shortages. Rhythms of stomatal opening and closure may ensure that the above-ground parts of the plant remain more or less watertight except during controlled periods of active photosynthesis. These rhythms may •••• P er ce nt ag e ca no py fu lln es s J 0 J 20 60 100 Month (a) 40 80 F M A M J J A S O N D P re da w n w at er p ot en tia l M P a) J –2.0 J 0.0 Month (b) –1.0 –0.5 F M A M J J A S O N D –1.5 A ss im ila tio n ra te ( µm ol m –2 s –1 ) J 0 J 2 12 16 Month (c) 10 14 F M A M J J A S O N D 4 6 8 Figure 3.10 (a) Percentage canopy fullness for deciduous (), semideciduous (), brevideciduous () and evergreen (
) trees Australian savannas throughout the year. (Note that the southern hemisphere dry season runs from around April to November.) (b) Susceptibility to drought as measured by increasingly negative values of ‘predawn water potential’ for deciduous () and evergreen (
) trees. (c) Net photosynthesis as measured by the carbon assimilation rate for deciduous () and evergreen (
) trees. (After Eamus, 1999.) leaf appearance and structure physiological strategies coexisting alternative strategies in Australian savannas EIPC03 10/24/05 1:47 PM Page 68 RESOURCES 69 be diurnal or may be quickly responsive to the plant’s internal water status. Stomatal movement may even be triggered directly by conditions at the leaf surface itself – the plant then responds to desiccating conditions at the very site, and at the same time, as the conditions are first sensed. 3.3 Carbon dioxide The CO2 used in photosynthesis is obtained almost entirely from the atmo- sphere, where its concentration has risen from approximately 280 µl l−1 in 1750 to about 370 µl l−1 today and is still increasing by 0.4–0.5% year−1 (see Figure 18.22). In a terrestrial community, the flux of CO2 at night is upwards, from the soil and vegetation to the atmosphere; on sunny days above a photosynthesizing canopy, there is a downward flux. Above a vegetation canopy, the air becomes rapidly mixed. However, the situation is quite different within and beneath canopies. Changes in CO2 con- centration in the air within a mixed deciduous forest in New England were measured at various heights above ground level during the year (Figure 3.11a) (Bazzaz & Williams, 1991). Highest concentrations, up to around 1800 µl l−1, were measured near the surface of the ground, tapering off to around 400 µl l−1 at 1 m above the ground. These high values near ground level were achieved in the summer when high temperatures allowed the rapid decomposition of litter and soil organic matter. At greater heights within the forest, the CO2 concentrations scarcely ever (even in winter) reached the value of 370 µl l−1 which is the atmospheric concentration of bulk air measured at the Mauna Loa laboratory in Hawaii (see Figure 18.22). During the winter months, concentrations remained virtually constant through the day and night at all heights. But in the summer, major diurnal cycles of concentration developed that reflected the interaction between the production of CO2 by decomposition and its con- sumption in photosynthesis (Figure 3.11b). That CO2 concentrations vary so widely within vegetation means that plants growing in different parts of a forest will experience quite different CO2 environments. Indeed the lower leaves on a forest shrub will usually experience higher CO2 concentrations than its upper leaves, and seedlings will live in environments richer in CO2 than mature trees. In aquatic environments, variations in CO2 concentration can be just as striking, especially when water mixing is limited, for example during the sum- mer ‘stratification’ of lakes, with layers of warm water towards the surface and colder layers beneath (Figure 3.12). •••• C O 2 co nc en tr at io ns ( µl l– 1 ) Dec 31Nov 11Apr 25Mar 6 440 Sep 22 Measurement date (a) 420 400 380 360 340 320 300 Jun 14 Aug 3 Time of day 0400 255 455 405 355 305 20001200 Jul 4 C O 2 co nc en tr at io ns ( µl l– 1 ) 0400 255 (b) 455 405 355 305 20001200 Nov 21 Figure 3.11 (a) CO2 concentrations in a mixed deciduous forest (Harvard Forest, Massachusetts, USA) at various times of year at five heights above ground: , 0.05 m; 4, 0.20 m; , 3.00 m; 7, 6.00 m;
, 12.00 m. Data from the Mauna Loa CO2 observatory (5) are given on the same axis for comparison. (b) CO2 concentrations for each hour of the day (averaged over 3–7-day periods) on November 21 and July 4. (After Bazzaz & Williams, 1991.) the rise in global levels variations beneath a canopy variations in aquatic habitats . . . EIPC03 10/24/05 1:47 PM Page 69 72 CHAPTER 3 and the clearing of forests. As Loladze (2002) points out, while consequential changes to global climate may be controversial in some quarters, marked increases in CO2 concentration itself are not. Plants now are experiencing around a 30% higher concentra- tion compared to the pre-industrial period – effectively instantan- eous on geological timescales; trees living now may experience a doubling in concentration over their lifetimes – effectively an instantaneous change on an evolutionary timescale; and high mixing rates in the atmosphere mean that these are changes that will affect all plants. There is also evidence of large- scale changes in atmospheric CO2 over much longer timescales. Carbon balance models suggest that during the Triassic, Jurassic and Cretaceous periods, atmospheric concen- trations of CO2 were four to eight times greater than at present, falling after the Cretaceous from between 1400 and 2800 µl l−1 to below 1000 µl l−1 in the Eocene, Miocene and Pliocene, and fluctuating between 180 and 280 µl l−1 during subsequent glacial and interglacial periods (Ehleringer & Monson, 1993). The declines in CO2 concentration in the atmosphere after the Cretaceous may have been the primary force that favored the evo- lution of plants with C4 physiology (Ehleringer et al., 1991), because at low concentrations of CO2, photorespiration places C3 plants at a particular disadvantage. The steady rise in CO2 since the Industrial Revolution is therefore a partial return to pre- Pleistocene conditions and C4 plants may begin to lose some of their advantage. •••• (a) 0.37 0.00 1.40 1.50 0.45 0.56 1.34 0.99 2.13 1.77 2.84 3.20 3.36 4.38 2.04 1.77 0.69 0.17 0.24 0.38 0.41 0.080.29 0.81 0.81 0.38 0.72 0.56 0.00 0.00 0.00 0.00 0.00 0.15 0.28 0.43 2.54 0.31 0.22 C 4 sp ec ie s (% ) 806535 0 20 1 2 50 Mean summer pan evaporation (inches per summer) (b) 4 3 r = 0.947 Figure 3.14 (a) The percentage of native C4 dicot species in various regions of North America. (b) The relationship between the percentage of native C4 species in 31 geographic regions of North America, and the mean summer (May–October) pan evaporation – a climatic indicator of plant/water balance. Regions for which appropriate climatic data were unavailable were excluded, together with south Florida, where the peculiar geography and climate may explain the aberrant composition of the flora. (After Stowe & Teeri, 1978.) changes in geological time EIPC03 10/24/05 1:47 PM Page 72 ps [eta apito] as a 7 É Ecology From Individuals to Ecosystems Michael Begon, Colin R. Townsend, and John L. Harper •• ECOLOGY From Individuals to Ecosystems EIPA01 10/24/05 1:36 PM Page i •••• © 1986, 1990, 1996, 2006 by Blackwell Publishing Ltd BLACKWELL PUBLISHING 350 Main Street, Malden, MA 02148-5020, USA 9600 Garsington Road, Oxford OX4 2DQ, UK 550 Swanston Street, Carlton, Victoria 3053, Australia The right of Mike Begon, Colin Townsend and John Harper to be identified as the Authors of this Work has been asserted in accordance with the UK Copyright, Designs and Patents Act 1988. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by the UK Copyright, Designs, and Patents Act 1988, without the prior permission of the publisher First edition published 1986 by Blackwell Publishing Ltd Second edition published 1990 Third edition published 1996 Fourth edition published 2006 1 2006 Library of Congress Cataloging-in-Publication Data Begon, Michael. Ecology : from individuals to ecosystems / Michael Begon, Colin R. Townsend, John L. Harper.—4th ed. p. cm. Includes bibliographical references and index. ISBN-13: 978-1-4051-1117-1 (hard cover : alk. paper) ISBN-10: 1-4051-1117-8 (hard cover : alk. paper) 1. Ecology. I. Townsend, Colin R. II. Harper, John L. III. Title. QH54.B416 2005 577—dc22 2005004136 A catalogue record for this title is available from the British Library. Set in 9.5/12 Dante MT by Graphicraft Limited, Hong Kong Printed and bound in the United Kingdom by CPI Bath Press The publisher’s policy is to use permanent paper from mills that operate a sustainable forestry policy, and which has been manufatured from pulp processed using acid-free and elementary chlorine-free practices. Furthermore, the publisher ensures that the text paper and cover board used have met acceptable environmental accreditation standards. For further information on Blackwell Publishing, visit our website: www.blackwellpublishing.com EIPA01 10/24/05 1:36 PM Page iv •••• Contents Preface, vii Introduction: Ecology and its Domain, xi Part 1: Organisms 1 Organisms in their Environments: the Evolutionary Backdrop, 3 2 Conditions, 30 3 Resources, 58 4 Life, Death and Life Histories, 89 5 Intraspecific Competition, 132 6 Dispersal, Dormancy and Metapopulations, 163 7 Ecological Applications at the Level of Organisms and Single-Species Populations: Restoration, Biosecurity and Conservation, 186 Part 2: Species Interactions 8 Interspecific Competition, 227 9 The Nature of Predation, 266 10 The Population Dynamics of Predation, 297 11 Decomposers and Detritivores, 326 12 Parasitism and Disease, 347 13 Symbiosis and Mutualism, 381 14 Abundance, 410 15 Ecological Applications at the Level of Population Interactions: Pest Control and Harvest Management, 439 EIPA01 10/24/05 1:36 PM Page v •••• vi CONTENTS Part 3: Communities and Ecosystems 16 The Nature of the Community: Patterns in Space and Time, 469 17 The Flux of Energy through Ecosystems, 499 18 The Flux of Matter through Ecosystems, 525 19 The Influence of Population Interactions on Community Structure, 550 20 Food Webs, 578 21 Patterns in Species Richness, 602 22 Ecological Applications at the Level of Communities and Ecosystems: Management Based on the Theory of Succession, Food Webs, Ecosystem Functioning and Biodiversity, 633 References, 659 Organism Index, 701 Subject Index, 714 Color plate section between pp. 000 and 000 EIPA01 10/24/05 1:36 PM Page vi PREFACE ix collaborated with John over so many years, but also just how much we learnt from him. We cannot promise to have absorbed or, to be frank, to have accepted, every one of his views; and we hope in particular, in this fourth edition, that we have not strayed too far from the paths through which he has guided us. But if readers recognize any attempts to stimulate and inspire rather than simply to inform, to question rather than to accept, to respect our readers rather than to patronize them, and to avoid unques- tioning obedience to current reputation while acknowledging our debt to the masters of the past, then they will have identified John’s intellectual legacy still firmly imprinted on the text. In previous editions we thanked the great many friends and colleagues who helped us by commenting on various drafts of the text. The effects of their contributions are still strongly evident in the present edition. This fourth edition was also read by a series of reviewers, to whom we are deeply grateful. Several remained anonymous and so we cannot thank them by name, but we are delighted to be able to acknowledge the help of Jonathan Anderson, Mike Bonsall, Angela Douglas, Chris Elphick, Valerie Eviner, Andy Foggo, Jerry Franklin, Kevin Gaston, Charles Godfray, Sue Hartley, Marcel Holyoak, Jim Hone, Peter Hudson, Johannes Knops, Xavier Lambin, Svata Louda, Peter Morin, Steve Ormerod, Richard Sibly, Andrew Watkinson, Jacob Weiner, and David Wharton. At Blackwell, and in the production stage, we were particularly helped and encouraged by Jane Andrew, Elizabeth Frank, Rosie Hayden, Delia Sandford and Nancy Whilton. This book is dedicated to our families – by Mike to Linda, Jessica and Robert, and by Colin to Laurel, Dominic, Jenny and Brennan, and especially to the memory of his mother, Jean Evelyn Townsend. Mike Begon Colin Townsend •••• EIPA01 10/24/05 1:36 PM Page ix •••• EIPA01 10/24/05 1:36 PM Page x •••• Definition and scope of ecology The word ‘ecology’ was first used by Ernest Haeckel in 1869. Paraphrasing Haeckel we can describe ecology as the scientific study of the interactions between organisms and their environ- ment. The word is derived from the Greek oikos, meaning ‘home’. Ecology might therefore be thought of as the study of the ‘home life’ of living organisms. A less vague definition was suggested by Krebs (1972): ‘Ecology is the scientific study of the interactions that determine the distribution and abundance of organisms’. Notice that Krebs’ definition does not use the word ‘environment’; to see why, it is necessary to define the word. The environment of an organism consists of all those factors and phenomena outside the organism that influence it, whether these are physical and chemical (abiotic) or other organisms (biotic). The ‘interactions’ in Krebs’ definition are, of course, interactions with these very factors. The environment therefore retains the central position that Haeckel gave it. Krebs’ definition has the merit of pinpointing the ultimate subject matter of ecology: the distribu- tion and abundance of organisms – where organisms occur, how many occur there, and why. This being so, it might be better still to define ecology as: the scientific study of the distribution and abundance of organisms and the interactions that determine distribution and abundance. As far as the subject matter of ecology is concerned, ‘the distribution and abundance of organisms’ is pleasantly succinct. But we need to expand it. The living world can be viewed as a biological hierarchy that starts with subcellular particles, and continues up through cells, tissues and organs. Ecology deals with the next three levels: the individual organism, the population (consisting of individuals of the same species) and the community (consisting of a greater or lesser number of species populations). At the level of the organism, ecology deals with how individuals are affected by (and how they affect) their environment. At the level of the population, ecology is concerned with the presence or absence of particular species, their abundance or rarity, and with the trends and fluctuations in their numbers. Community ecology then deals with the composition and organization of ecological communities. Ecologists also focus on the pathways followed by energy and matter as these move among living and nonliving elements of a further category of organization: the ecosystem, comprising the community together with its physical environment. With this in mind, Likens (1992) would extend our preferred definition of ecology to include ‘the interactions between organisms and the transformation and flux of energy and matter’. However, we take energy/matter transformations as being subsumed in the ‘interactions’ of our definition. There are two broad approaches that ecologists can take at each level of ecological organization. First, much can be gained by building from properties at the level below: physiology when studying organismal ecology; individual clutch size and survival probabilities when investigating the dynamics of individual species populations; food consumption rates when dealing with inter- actions between predator and prey populations; limits to the similarity of coexisting species when researching communities, and so on. An alternative approach deals directly with properties of the level of interest – for example, niche breadth at the organis- mal level; relative importance of density-dependent processes at the population level; species diversity at the level of community; rate of biomass production at the ecosystem level – and tries to relate these to abiotic or biotic aspects of the environment. Both approaches have their uses, and both will be used in each of the three parts of this book: Organisms; Species Interactions; and Communities and Ecosystems. Introduction: Ecology and its Domain EIPA01 10/24/05 1:36 PM Page xi •• EIPC01 10/24/05 1:42 PM Page 2 •• 1.1 Introduction: natural selection and adaptation From our definition of ecology in the Preface, and even from a layman’s understanding of the term, it is clear that at the heart of ecology lies the relationship between organisms and their environments. In this opening chapter we explain how, funda- mentally, this is an evolutionary relationship. The great Russian– American biologist Theodosius Dobzhansky famously said: ‘Nothing in biology makes sense, except in the light of evolution’. This is as true of ecology as of any other aspect of biology. Thus, we try here to explain the processes by which the properties of different sorts of species make their life possible in particular environments, and also to explain their failure to live in other environments. In mapping out this evolutionary backdrop to the subject, we will also be introducing many of the questions that are taken up in detail in later chapters. The phrase that, in everyday speech, is most commonly used to describe the match between organisms and environment is: ‘organism X is adapted to’ followed by a description of where the organism is found. Thus, we often hear that ‘fish are adapted to live in water’, or ‘cacti are adapted to live in conditions of drought’. In everyday speech, this may mean very little: simply that fish have characteristics that allow them to live in water (and perhaps exclude them from other environments) or that cacti have characteristics that allow them to live where water is scarce. The word ‘adapted’ here says nothing about how the characteristics were acquired. For an ecologist or evolutionary biologist, however, ‘X is adapted to live in Y’ means that environment Y has provided forces of natural selection that have affected the life of X’s ancestors and so have molded and specialized the evolution of X. ‘Adaptation’ means that genetic change has occurred. Regrettably, though, the word ‘adaptation’ implies that organisms are matched to their present environments, suggest- ing ‘design’ or even ‘prediction’. But organisms have not been designed for, or fitted to the present: they have been molded (by natural selection) by past environments. Their characteristics reflect the successes and failures of ancestors. They appear to be apt for the environments that they live in at present only because present environments tend to be similar to those of the past. The theory of evolution by natural selection is an ecological theory. It was first elaborated by Charles Darwin (1859), though its essence was also appreciated by a contemporary and corres- pondent of Darwin’s, Alfred Russell Wallace (Figure 1.1). It rests on a series of propositions. 1 The individuals that make up a population of a species are not identical: they vary, although sometimes only slightly, in size, rate of development, response to temperature, and so on. 2 Some, at least, of this variation is heritable. In other words, the characteristics of an individual are determined to some extent by its genetic make-up. Individuals receive their genes from their ancestors and therefore tend to share their characteristics. 3 All populations have the potential to populate the whole earth, and they would do so if each individual survived and each indi- vidual produced its maximum number of descendants. But they do not: many individuals die prior to reproduction, and most (if not all) reproduce at a less than maximal rate. 4 Different ancestors leave different numbers of descendants. This means much more than saying that different individuals produce different numbers of offspring. It includes also the chances of survival of offspring to reproductive age, the survival and reproduction of the progeny of these offspring, the survival and reproduction of their offspring in turn, and so on. 5 Finally, the number of descendants that an individual leaves depends, not entirely but crucially, on the interaction between the characteristics of the individual and its environment. the meaning of adaptation evolution by natural selection Chapter 1 Organisms in their Environments: the Evolutionary Backdrop EIPC01 10/24/05 1:42 PM Page 3 4 CHAPTER 1 In any environment, some individuals will tend to survive and reproduce better, and leave more descendants, than others. If, because of this, the heritable characteristics of a population change from generation to generation, then evolution by nat- ural selection is said to have occurred. This is the sense in which nature may loosely be thought of as selecting. But nature does not select in the way that plant and animal breeders select. Breeders have a defined end in view – bigger seeds or a faster racehorse. But nature does not actively select in this way: it simply sets the scene within which the evolutionary play of differential survival and reproduction is played out. The fittest individuals in a popula- tion are those that leave the greatest number of descendants. In practice, the term is often applied not to a single individual, but to a typ- ical individual or a type. For example, we may say that in sand dunes, yellow-shelled snails are fitter than brown-shelled snails. Fitness, then, is a relative not an absolute term. The fittest indi- viduals in a population are those that leave the greatest number of descendants relative to the number of descendants left by other individuals in the population. When we marvel at the diversity of complex specializations, there is a temptation to regard each case as an example of evolved perfection. But this would be wrong. The evolutionary process works on the genetic variation that is avail- able. It follows that natural selection is unlikely to lead to the evolution of perfect, ‘maximally fit’ individuals. Rather, organisms •••• Figure 1.1 (a) Charles Darwin, 1849 (lithograph by Thomas H. Maguire; courtesy of The Royal Institution, London, UK/Bridgeman Art Library). (b) Alfred Russell Wallace, 1862 (courtesy of the Natural History Museum, London). fitness: it’s all relative evolved perfection? no (a) (b) EIPC01 10/24/05 1:42 PM Page 4 THE EVOLUTIONARY BACKDROP 7 the extent to which the genetic response of populations to environmental change will always be out of step with the environment and unable to anticipate changing circumstances – this is illustrated in the peppered moth example below. Many polymorphisms, however, are actively maintained in a population by natural selection, and there are a num- ber of ways in which this may occur. 1 Heterozygotes may be of superior fitness, but because of the mechanics of Mendelian genetics they continually generate less fit homozygotes within the population. Such ‘heterosis’ is seen in human sickle-cell anaemia where malaria is prevalent. The malaria parasite attacks red blood cells. The sickle-cell muta- tion gives rise to red cells that are physiologically imperfect and misshapen. However, sickle-cell heterozygotes are fittest because they suffer only slightly from anemia and are little affected by malaria; but they continually generate homozygotes that are either dangerously anemic (two sickle-cell genes) or susceptible to malaria (no sickle-cell genes). None the less, the superior fitness of the heterozygote maintains both types of gene in the population (that is, a polymorphism). 2 There may be gradients of selective forces favoring one form (morph) at one end of the gradient, and another form at the other. This can produce polymorphic populations at inter- mediate positions in the gradient – this, too, is illustrated below in the peppered moth study. 3 There may be frequency-dependent selection in which each of the morphs of a species is fittest when it is rarest (Clarke & Partridge, 1988). This is believed to be the case when rare color forms of prey are fit because they go unrecognized and are therefore ignored by their predators. 4 Selective forces may operate in different directions within different patches in the population. A striking example of this is provided by a reciprocal transplant study of white clover (Trifolium repens) in a field in North Wales (UK). To determine whether the characteristics of individuals matched local features of their environment, Turkington and Harper (1979) removed plants from marked positions in the field and multiplied them into clones in the common environment of a greenhouse. They then transplanted samples from each clone into the place in the sward of vegetation from which it had originally been taken (as a control), and also to the places from where all the others had been taken (a transplant). The plants were allowed to grow for a year before they were removed, dried and weighed. The mean weight of clover plants transplanted back into their home sites was 0.89 g but at away sites it was only 0.52 g, a statistically highly significant difference. This provides strong, direct evidence that clover clones in the pasture had evolved to become specialized such that they performed best in their local environment. But all this was going on within a single population, which was therefore polymorphic. In fact, the distinction between local ecotypes and polymorphic popu- lations is not always a clear one. This is illustrated by another study in North Wales, where there was a gradation in habitats at the margin between maritime cliffs and grazed pasture, and a common species, creeping bent grass (Agrostis stolonifera), was present in many of the habitats. Figure 1.4 shows a map of the site and one of the transects from which plants were sampled. It also shows the results when plants from the sampling points along this transect were grown in a common garden. The •••• Figure 1.4 (a) Map of Abraham’s Bosom, the site chosen for a study of evolution over very short distances. The darker colored area is grazed pasture; the lighter areas are the cliffs falling to the sea. The numbers indicate the sites from which the grass Agrostis stolonifera was sampled. Note that the whole area is only 200 m long. (b) A vertical transect across the study area showing the gradual change from pasture to cliff conditions. (c) The mean length of stolons produced in the experimental garden from samples taken from the transect. (From Aston & Bradshaw, 1966.) the maintenance of polymorphisms no clear distinction between local ecotypes and a polymorphism 1 2 3 4 5 N 0 200 m100 Irish Sea (a) 1 2 3 5 4 100 30 20 10 0 E le va tio n (m ) 0 (b) 100 50 25 0 S to lo n le ng th ( cm ) 0 (c) Distance (m) EIPC01 10/24/05 1:42 PM Page 7 8 CHAPTER 1 plants spread by sending out shoots along the ground surface (stolons), and the growth of plants was compared by measuring the lengths of these. In the field, cliff plants formed only short stolons, whereas those of the pasture plants were long. In the experi- mental garden, these differences were maintained, even though the sampling points were typically only around 30 m apart – certainly within the range of pollen dispersal between plants. Indeed, the gradually changing environment along the transect was matched by a gradually changing stolon length, presumably with a genetic basis, since it was apparent in the common garden. Thus, even though the spatial scale was so small, the forces of selection seem to outweigh the mixing forces of hybridization – but it is a moot point whether we should describe this as a small-scale series of local ecotypes or a polymorphic population maintained by a gradient of selection. 1.2.3 Variation within a species with manmade selection pressures It is, perhaps, not surprising that some of the most dramatic examples of local specialization within species (indeed of natural selection in action) have been driven by manmade ecological forces, especially those of environmental pollution. These can provide rapid change under the influence of powerful selection pressures. Industrial melanism, for example, is the phenomenon in which black or blackish forms of species have come to dominate populations in industrial areas. In the dark individuals, a dominant gene is typ- ically responsible for producing an excess of the black pigment melanin. Industrial melanism is known in most industrialized coun- tries and more than 100 species of moth have evolved forms of industrial melanism. •••• f. insularia f. carbonaria f. typica Figure 1.5 Sites in Britain where the frequencies of the pale ( forma typica) and melanic forms of Biston betularia were recorded by Kettlewell and his colleagues. In all more than 20,000 specimens were examined. The principal melanic form ( forma carbonaria) was abundant near industrial areas and where the prevailing westerly winds carry atmospheric pollution to the east. A further melanic form ( forma insularia, which looks like an intermediate form but is due to several different genes controlling darkening) was also present but was hidden where the genes for forma carbonaria were present. (From Ford, 1975.) EIPC01 10/24/05 1:42 PM Page 8 THE EVOLUTIONARY BACKDROP 9 The earliest recorded species to evolve in this way was the peppered moth (Biston betularia); the first black specimen in an otherwise pale popula- tion was caught in Manchester (UK) in 1848. By 1895, about 98% of the Manchester peppered moth popu- lation was melanic. Following many more years of pollution, a large-scale survey of pale and melanic forms of the peppered moth in Britain recorded more than 20,000 specimens between 1952 and 1970 (Figure 1.5). The winds in Britain are predominantly westerlies, spreading industrial pollutants (especially smoke and sulfur dioxide) toward the east. Melanic forms were concentrated toward the east and were completely absent from the unpolluted western parts of England and Wales, northern Scotland and Ireland. Notice from the figure, though, that many populations were polymorphic: melanic and nonmelanic forms coexisted. Thus, the polymorphism seems to be a result both of environ- ments changing (becoming more polluted) – to this extent the poly- morphism is transient – and of there being a gradient of selective pressures from the less polluted west to the more polluted east. The main selective pressure appears to be applied by birds that prey on the moths. In field experiments, large numbers of melanic and pale (‘typical’) moths were reared and released in equal numbers. In a rural and largely unpolluted area of southern England, most of those captured by birds were melanic. In an industrial area near the city of Birmingham, most were typicals (Kettlewell, 1955). Any idea, however, that melanic forms were favored simply because they were camouflaged against smoke- stained backgrounds in the polluted areas (and typicals were favored in unpolluted areas because they were camouflaged against pale backgrounds) may be only part of the story. The moths rest on tree trunks during the day, and nonmelanic moths are well hidden against a background of mosses and lichens. Industrial pollution has not just blackened the moths’ background; sulfur dioxide, especially, has also destroyed most of the moss and lichen on the tree trunks. Thus, sulfur dioxide pollution may have been as important as smoke in selecting melanic moths. In the 1960s, industrialized environments in Western Europe and the United States started to change again, as oil and electricity began to replace coal, and legislation was passed to impose smoke- free zones and to reduce industrial emissions of sulfur dioxide. The frequency of melanic forms then fell back to near pre- Industrial levels with remarkable speed (Figure 1.6). Again, there was transient polymorphism – but this time while populations were en route in the other direction. 1.3 Speciation It is clear, then, that natural selection can force populations of plants and animals to change their character – to evolve. But none of the examples we have considered has involved the evolution of a new species. What, then, justifies naming two populations as different species? And what is the process – ‘speciation’ – by which two or more new species are formed from one original species? 1.3.1 What do we mean by a ‘species’? Cynics have said, with some truth, that a species is what a competent taxonomist regards as a species. On the other hand, back in the 1930s two American biologists, Mayr and Dobzhansky, proposed an empir- ical test that could be used to decide whether two populations were part of the same species or of two different species. They recognized organisms as being members of a single species if they could, at least potentially, breed together in nature to produce fertile offspring. They called a species tested and defined in this way a biological species or biospecies. In the examples that we have used earlier in this chapter we know that melanic and normal peppered moths can mate and that the offspring are fully fertile; this is also true of plants from the different types of Agrostis. They are all variations within species – not separate species. In practice, however, biologists do not apply the Mayr– Dobzhansky test before they recognize every species: there is simply not enough time or resources, and in any case, there are vast portions of the living world – most microorganisms, for example – where an absence of sexual reproduction makes a strict interbreeding criterion inappropriate. What is more important is that the test recognizes a crucial element in the evolutionary process that we have met already in considering specialization •••• industrial melanism in the peppered moth 100 80 60 40 20 0 F re qu en cy 1950 1960 1970 Year 1980 1990 2000 Figure 1.6 Change in the frequency of the carbonaria form of the peppered moth Biston betularia in the Manchester area since 1950. Vertical lines show the standard error and the horizontal lines show the range of years included. (After Cook et al., 1999.) biospecies: the Mayr– Dobzhansky test EIPC01 10/24/05 1:42 PM Page 9 •• 12 CHAPTER 1 •• 14 g 20 g 34 g 21 g 28 g 20 g 13 g 20 g 18 g 21 g 34 g 8 g 13 g 10 g G. fuliginosa G. fortis G. magnirostris G. scandens G. conirostris G. difficilis C. parvulus C. psittacula C. pauper C. pallida P. crassirostris Ce. fusca Pi. inornata Ce. olivacea Scratch for seeds on the ground Feed on seeds on the ground and the flowers and pulp of prickly pear (Opuntia) Feed in trees on beetles Use spines held in the bill to extract insects from bark crevices Feed on leaves, buds and seeds in the canopy of trees Warbler-like birds feeding on small soft insects (b) 10°N 5°N 0° 90°W 85°W 80°W Culpepper Wenman Pinta Galapágos Santa Cruz San Cristobal HoodIsabela Fernandina Cocos Island Pearl Is. (a) Figure 1.9 (a) Map of the Galápagos Islands showing their position relative to Central America; on the equator 5° equals approximately 560 km. (b) A reconstruction of the evolutionary history of the Galápagos finches based on variation in the length of microsatellite deoxyribonucleic acid (DNA). The feeding habits of the various species are also shown. Drawings of the birds are proportional to actual body size. The maximum amount of black coloring in male plumage and the average body mass are shown for each species. The genetic distance (a measure of the genetic difference) between species is shown by the length of the horizontal lines. Notice the great and early separation of the warbler finch (Certhidea olivacea) from the others, suggesting that it may closely resemble the founders that colonized the islands. C, Camarhynchus; Ce, Certhidea; G, Geospiza; P, Platyspiza; Pi, Pinaroloxias. (After Petren et al., 1999.) EIPC01 10/24/05 1:42 PM Page 12 •• THE EVOLUTIONARY BACKDROP 13 psittacula) has become insectivorous, feeding on beetles and other insects in the canopy of trees. A so-called woodpecker finch, Camarhynchus (Cactospiza) pallida, extracts insects from crevices by holding a spine or a twig in its bill, while yet a fur- ther group includes the warbler finch, which flits around actively and collects small insects in the forest canopy and in the air. Isolation – both of the archipelago itself and of individual islands within it – has led to an original evolutionary line radiating into a series of species, each matching its own environment. 1.4 Historical factors Our world has not been constructed by someone taking each species in turn, testing it against each environment, and molding it so that every species finds its perfect place. It is a world in which species live where they do for reasons that are often, at least in part, accidents of history. We illustrate this first by continuing our examination of islands. 1.4.1 Island patterns Many of the species on islands are either subtly or profoundly dif- ferent from those on the nearest comparable area of mainland. Put simply, there are two main reasons for this. 1 The animals and plants on an island are limited to those types having ancestors that managed to disperse there, although the extent of this limitation depends on the isolation of the island and the intrinsic dispersal ability of the animal or plant in question. 2 Because of this isolation, as we saw in the previous section, the rate of evolutionary change on an island may often be fast enough to outweigh the effects of the exchange of genetic material between the island population and related populations elsewhere. Thus, islands contain many species unique to themselves (‘endemics’ – species found in only one area), as well as many differentiated ‘races’ or ‘subspecies’ that are distinguishable from mainland forms. A few individuals that disperse by chance to a habitable island can form the nucleus of an expanding new species. Its character will have been colored by the particular genes that were represented among the colonists – which are unlikely to be a perfect sample of the parent population. What natural selection can do with this founder population is limited by what is in its limited sample of genes (plus occasional rare mutations). Indeed much of the deviation among populations isolated on islands appears to be due to a founder effect – the chance composition of the pool of founder genes puts limits and constraints on what variation there is for natural selection to act upon. The Drosophila fruit-flies of Hawaii provide a further spec- tacular example of species formation on islands. The Hawaiian chain of islands (Figure 1.10) is volcanic in origin, having been formed gradually over the last 40 million years, as the center of the Pacific tectonic plate moved steadily over a ‘hot spot’ in a southeasterly direction (Niihau is the most ancient of the islands, Hawaii itself the most recent). The richness of the Hawaiian Drosophila is spectacular: there are probably about 1500 Drosophila spp. worldwide, but at least 500 of these are found only in the Hawaiian islands. Of particular interest are the 100 or so species of ‘picture-winged’ Droso- phila. The lineages through which these species have evolved can be traced by analyzing the banding patterns on the giant chro- mosomes in the salivary glands of their larvae. The evolutionary tree that emerges is shown in Figure 1.10, with each species lined up above the island on which it is found (there are only two species found on more than one island). The historical element in ‘what lives where’ is plainly apparent: the more ancient species live on the more ancient islands, and, as new islands have been formed, rare dispersers have reached them and eventually evolved in to new species. At least some of these species appear to match the same environment as others on different islands. Of the closely related species, for example, D. adiastola (species 8) is only found on Maui and D. setosimentum (species 11) only on Hawaii, but the environments that they live in are apparently indistinguishable (Heed, 1968). What is most noteworthy, of course, is the power and importance of isolation (coupled with natural selection) in generating new species. Thus, island biotas illustrate two import- ant, related points: (i) that there is a historical element in the match between organisms and environments; and (ii) that there is not just one perfect organism for each type of environment. 1.4.2 Movements of land masses Long ago, the curious distributions of species between continents, seemingly inexplicable in terms of dispersal over vast distances, led biologists, especially Wegener (1915), to suggest that the continents themselves must have moved. This was vigorously denied by geologists, until geomagnetic measurements required the same, apparently wildly improbable explanation. The discovery that the tectonic plates of the earth’s crust move and carry with them the migrating continents, reconciles geologist and biologist (Figure 1.11b–e). Thus, whilst major evolutionary developments were occurring in the plant and animal kingdoms, populations were being split and separated, and land areas were moving across climatic zones. Figure 1.12 shows just one example of a major group of organisms (the large flightless birds), whose distributions begin to make sense only in the light of the movement of land masses. It would be •• Hawaiian Drosophila large flightless birds EIPC01 10/24/05 1:42 PM Page 13 •••• 14 CHAPTER 1 N 62 95 68 70 54 53 43 55 85 86 76 99 81 91 77 84 89 75 59 60 61 67 74 69 83 82 97 90 94 81 50 52 49 51 48 3735 36 38 39 47 44 46 66 58 8180 98 punalua group (58–65) glabriapex group (34–57) grimshawi group (66–101) planitidia group (17–33) 40 41 42 2221 2524 26 27 23 18 19 17 20 34 32 1613 1514 6 4 5 1 adiastola group (3–16) 2 3 Niihau Kauai Oahu Lanai Molokai Maui Kahoolawe Hawaii 63 64 65 71 72 73 78 79 87 88 92 93 96 100 101 57 56 45 33 31 30 29 28 10 8 97 12 11 0 50 km Figure 1.10 An evolutionary tree linking the picture-winged Drosophila of Hawaii, traced by the analysis of chromosomal banding patterns. The most ancient species are D. primaeva (species 1) and D. attigua (species 2), found only on the island of Kauai. Other species are represented by solid circles; hypothetical species, needed to link the present day ones, are represented by open circles. Each species has been placed above the island or islands on which it is found (although Molokai, Lanai and Maui are grouped together). Niihau and Kahoolawe support no Drosophila. (After Carson & Kaneshiro, 1976; Williamson, 1981.) EIPC01 10/24/05 1:42 PM Page 14 THE EVOLUTIONARY BACKDROP 17 climate during the Pleistocene ice ages, in particular, bear a lot of the responsibility for the present patterns of distribution of plants and animals. The extent of these climatic and biotic changes is only beginning to be unraveled as the technology for discover- ing, analyzing and dating biological remains becomes more sophisticated (particularly by the analysis of buried pollen sam- ples). These methods increasingly allow us to determine just how much of the present distribution of organisms represents a precise local match to present environments, and how much is a fingerprint left by the hand of history. Techniques for the measurement of oxygen isotopes in ocean cores indic- ate that there may have been as many as 16 glacial cycles in the Pleistocene, each lasting for about 125,000 years (Figure 1.13a). It seems that each glacial phase may have lasted for as long as 50,000–100,000 years, with brief intervals of 10,000–20,000 years when the tem- peratures rose close to those we experience today. This suggests that it is present floras and faunas that are unusual, because they have developed towards the end of one of a series of unusual catas- trophic warm events! During the 20,000 years since the peak of the last glaciation, global temperatures have risen by about 8°C, and the rate at which vegetation has changed over much of this period has been detected by examining pollen records. The woody species that dominate pollen profiles at Rogers Lake in Connecticut (Figure 1.13b) have arrived in turn: spruce first and chestnut most recently. Each new arrival has added to the number of the species present, which has increased continually over the past 14,000-year period. The same picture is repeated in European profiles. As the number of pollen records has increased, it has become possible not only to plot the changes in vegetation •••• Te m pe ra tu re ( °C ) Time (103 years ago) 30 0 50 150 200 250 (a) 20 100 300 350 400 0 (b) 2 4 6 8 10 12 14 0 0 0 0 0 10,000 0 0 500 0 0 0 500 2000 1000 3000 10,000 20,000 2000 4000 1000 2000 5000 15,000 500 2000 1000 1000 2000 1000 Chestnut Hickory Beech Hemlock Oak Pine Pine Spruce Pi ce a Sp ru ce Pi nu s Pi ne Be tu la Bi rc h Ts ug a He m lo ck Q ue rc us O ak Ac er s ac ch ar um Su ga r m ap le Ac er ru br um Re d m ap le Fa gu s Be ec h Ca ry a Hi ck or y Ca st an ea Ch es tn ut 10 3 ye ar s ag o Figure 1.13 (a) An estimate of the temperature variations with time during glacial cycles over the past 400,000 years. The estimates were obtained by comparing oxygen isotope ratios in fossils taken from ocean cores in the Caribbean. The dashed line corresponds to the ratio 10,000 years ago, at the start of the present warming period. Periods as warm as the present have been rare events, and the climate during most of the past 400,000 years has been glacial. (After Emiliani, 1966; Davis, 1976.) (b) The profiles of pollen accumulated from late glacial times to the present in the sediments of Rogers Lake, Connecticut. The estimated date of arrival of each species in Connecticut is shown by arrows at the right of the figure. The horizontal scales represent pollen influx: 103 grains cm−2 year−1. (After Davis et al., 1973.) the Pleistocene glacial cycles . . . . . . from which trees are still recovering EIPC01 10/24/05 1:42 PM Page 17 18 CHAPTER 1 at a point in space, but to begin to map the movements of the various species as they have spread across the continents (see Bennet, 1986). In the invasions that followed the retreat of the ice in eastern North America, spruce was followed by jack pine or red pine, which spread northwards at a rate of 350–500 m year−1 for several thousands of years. White pine started its migration about 1000 years later, at the same time as oak. Hemlock was also one of the rapid invaders (200–300 m year−1), and arrived at most sites about 1000 years after white pine. Chestnut moved slowly (100 m year−1), but became a dominant species once it had arrived. Forest trees are still migrating into deglaciated areas, even now. This clearly implies that the timespan of an average interglacial period is too short for the attainment of floristic equilibrium (Davis, 1976). Such historical factors will have to be borne in mind when we consider the various patterns in species richness and biodiversity in Chapter 21. ‘History’ may also have an impact on much smaller space and time scales. Disturbances to the benthic (bottom dwelling) community of a stream occurs when high discharge events (associated with storms or snow melt) result in a very small-scale mosaic of patches of scour (substrate loss), fill (addition of substrate) and no change (Matthaei et al., 1999). The invertebrate communities associated with the differ- ent patch histories are distinctive for a period of months, within which time another high discharge event is likely to occur. As with the distribution of trees in relation to repeating ice ages, the stream fauna may rarely achieve an equilibrium between flow disturbances (Matthaei & Townsend, 2000). The records of climatic change in the tropics are far less complete than those for temperate regions. There is therefore the temptation to imagine that whilst dramatic climatic shifts and ice invasions were dom- inating temperate regions, the tropics persisted in the state we know today. This is almost certainly wrong. Data from a variety of sources indicate that there were abrupt fluctuations in post- glacial climates in Asia and Africa. In continental monsoon areas (e.g. Tibet, Ethiopia, western Sahara and subequatorial Africa) the postglacial period started with an extensive phase of high humid- ity followed by a series of phases of intense aridity (Zahn, 1994). In South America, a picture is emerging of vegetational changes that parallel those occurring in temperate regions, as the extent of tropical forest increased in warmer, wetter periods, and con- tracted, during cooler, drier glacial periods, to smaller patches surrounded by a sea of savanna. Support for this comes from the present-day distribution of species in the tropical forests of South America (Figure 1.14). There, particular ‘hot spots’ of species diversity are apparent, and these are thought to be likely sites of forest refuges during the glacial periods, and sites too, there- fore, of increased rates of speciation (Prance, 1987; Ridley, 1993). On this interpretation, the present distributions of species may again be seen as largely accidents of history (where the refuges were) rather than precise matches between species and their dif- fering environments. Evidence of changes in vegetation that followed the last retreat of the ice hint at the consequence of the global warming (maybe 3°C in the next 100 years) that is predicted to result from continuing increases in atmospheric carbon dioxide (discussed in detail in Sections 2.9.1 and 18.4.6). But the scales are quite different. Postglacial warming of about 8°C occurred over 20,000 years, and changes in the vegetation failed to keep pace even with this. But current projections for the 21st century require range shifts for trees at rates of 300–500 km per century compared to typical rates in the past of 20–40 km per century (and exceptional rates of 100–150 km). It is striking that the only pre- cisely dated extinction of a tree species in the Quaternary, that of Picea critchfeldii, occurred around 15,000 years ago at a time of especially rapid postglacial warming ( Jackson & Weng, 1999). Clearly, even more rapid change in the future could result in extinc- tions of many additional species (Davis & Shaw, 2001). •••• Napo Madiera Peru East Imeri Guiana (b)(a) Figure 1.14 (a) The present-day distribution of tropical forest in South America. (b) The possible distribution of tropical forest refuges at the time when the last glaciation was at its peak, as judged by present-day hot spots of species diversity within the forest. (After Ridley, 1993.) ‘history’ on a smaller scale changes in the tropics how will global warming compare? EIPC01 10/24/05 1:42 PM Page 18 THE EVOLUTIONARY BACKDROP 19 1.4.4 Convergents and parallels A match between the nature of organ- isms and their environment can often be seen as a similarity in form and behavior between organisms living in a similar environment, but belonging to different phyletic lines (i.e. different branches of the evolutionary tree). Such similarities also undermine further the idea that for every environment there is one, and only one, perfect organism. The evidence is particularly persuasive when the phyletic lines are far removed from each other, and when similar roles are played by structures that have quite different evolutionary origins, i.e. when the structures are analogous (similar in superficial form or function) but not homologous (derived from an equivalent structure in a common ancestry). When this is seen to occur, we speak of convergent evolution. Many flowering plants and some ferns, for example, use the support of others to climb high in the canopies of vegetation, and so gain access to more light than if they depended on their own supporting tissues. The ability to climb has evolved in many dif- ferent families, and quite different organs have become modified into climbing structures (Figure 1.15a): they are analogous struc- tures but not homologous. In other plant species the same organ has been modified into quite different structures with quite dif- ferent roles: they are therefore homologous, although they may not be analogous (Figure 1.15b). Other examples can be used to show the parallels in evolutionary pathways within separate groups that have radiated after they were isolated from each other. The classic example of such parallel evolution is the radiation amongst the placental and marsupial mammals. Marsupials arrived on the Australian continent in the Cretaceous period (around 90 million years ago), when the only other mammals present were the curious egg-laying monotremes (now represented only by the spiny anteaters (Tachyglossus aculeatus) and the duckbill platypus (Ornithorynchus anatinus)). An evolutionary process of radiation then occurred that in many •••• Dioscorea (Dioscoreaceae), twiner Calamus (Arecaceae), hooks Clematis (Ranunculaceae), twining petiole Cobaea (Cobaeaceae), tendril Ficus (Moraceae), adventitious roots Parthenocissus (Vitaceae), sticky pads (a) analogous and homologous structures Figure 1.15 A variety of morphological features that allow flowering plants to climb. (a) Structural features that are analogous, i.e. derived from modifications of quite different organs, e.g. leaves, petioles, stems, roots and tendrils. EIPC01 10/24/05 1:42 PM Page 19 22 CHAPTER 1 from the Low Arctic (around 600 species in North America) to the High Arctic (north of 83°, e.g. around 100 species in Greenland and Ellesmere Island). In contrast, the flora of Antarctica contains only two native species of vascular plant and some lichens and mosses that support a few small invertebrates. The biological productivity and diversity of Antarctica are con- centrated at the coast and depend almost entirely on resources harvested from the sea. Taiga or northern coniferous forest (see Plate 1.2, facing p. XX) occupies a broad belt across North America and Eurasia. Liquid water is unavailable for much of the winter, and plants and many of the animals have a conspicuous winter dor- mancy in which metabolism is very slow. Generally, the tree flora is very limited. In areas with less severe winters, the forests may be dominated by pines (Pinus species, which are all evergreens) and deciduous trees such as larch (Larix), birch (Betula) or aspens (Populus), often as mixtures of species. Farther north, these species give way to single-species forests of spruce (Picea) cover- ing immense areas. The overriding environmental constraint in northern spruce forests is the presence of permafrost, creating drought except when the sun warms the surface. The root system of spruce can develop in the superficial soil layer, from which the trees derive all their water during the short growing season. Temperate forests (see Plate 1.3, between pp. XX and XX) range from the mixed conifer and broad-leaved forests of much of North America and northern central Europe (where there may be 6 months of freezing temperatures), to the moist dripping forests of broad-leaved evergreen trees found at the biome’s low latitude limits in, for example, Florida and New Zealand. In most temperate forests, however, there are periods of the year when liquid water is in short supply, because poten- tial evaporation exceeds the sum of precipitation and water available from the soil. Deciduous trees, which dominate in most temperate forests, lose their leaves in the fall and become dormant. On the forest floor, diverse floras of perennial herbs often occur, particularly those that grow quickly in the spring before the new tree foliage has developed. Temperate forests also •••• Arctic tundra Northern coniferous forest Temperate forest Tropical rainforest Tropical seasonal forest Temperate grassland Tropical savanna grassland and scrub Desert Mediterranean vegetation, chaparral Mountains Figure 1.17 World distribution of the major biomes of vegetation. (After Audesirk & Audesirk, 1996.) taiga temperate forests EIPC01 10/24/05 1:42 PM Page 22 THE EVOLUTIONARY BACKDROP 23 provide food resources for animals that are usually very seasonal in their occurrence. Many of the birds of temperate forests are migrants that return in spring but spend the remainder of the year in warmer biomes. Grassland occupies the drier parts of temperate and tropical regions. Temperate grassland has many local names: the steppes of Asia, the prairies of North America, the pampas of South America and the veldt of South Africa. Tropical grassland or savanna (see Plate 1.4, between pp. XX and XX) is the name applied to tropical vegetation ranging from pure grass- land to some trees with much grass. Almost all of these temper- ate and tropical grasslands experience seasonal drought, but the role of climate in determining their vegetation is almost completely overridden by the effects of grazing animals that limit the species present to those that can recover from frequent defoliation. In the savanna, fire is also a common hazard in the dry season and, like grazing animals, it tips the balance in the vegetation against trees and towards grassland. None the less, there is typically a sea- sonal glut of food, alternating with shortage, and as a consequence the larger grazing animals suffer extreme famine (and mortality) in drier years. A seasonal abundance of seeds and insects supports large populations of migrating birds, but only a few species can find sufficiently reliable resources to be resident year-round. Many of these natural grasslands have been cultivated and replaced by arable annual ‘grasslands’ of wheat, oats, barley, rye and corn. Such annual grasses of temperate regions, together with rice in the tropics, provide the staple food of human popu- lations worldwide. At the drier margins of the biome, many of the grasslands are ‘managed’ for meat or milk production, some- times requiring a nomadic human lifestyle. The natural popula- tions of grazing animals have been driven back in favor of cattle, sheep and goats. Of all the biomes, this is the one most coveted, used and transformed by humans. Chaparral or maquis occurs in Mediterranean-type climates (mild, wet winters and summer drought) in Europe, California and northwest Mexico, and in a few small areas in Australia, Chile and South Africa. Chaparral develops in regions with less rainfall than temperate grasslands and is dominated mainly by a •••• 40 –60 0 5000 40 –60 0 5000 (b) Savanna 40 –60 0 M in im um te m pe ra tu re (m on th ly a ve ra ge ,° C ) M in im um te m pe ra tu re (m on th ly a ve ra ge ,° C ) 5000 (c) Temperate deciduous forest (d) Northern coniferous forest (taiga) (e) Tundra Total annual rainfall (mm) (a) Tropical rainforest 40 –60 0 5000 40 –60 0 5000 M in im um te m pe ra tu re (m on th ly a ve ra ge ,° C ) Congo (Africa) Manaus (South America) Atherton (Australia) Figure 1.18 The variety of environmental conditions experienced in terrestrial environments can be described in terms of their annual rainfall and mean monthly minimum temperatures. The range of conditions experienced in: (a) tropical rainforest, (b) savanna, (c) temperate deciduous forest, (d) northern coniferous forest (taiga), and (e) tundra. (After Heal et al., 1993; © UNESCO.) grassland chaparral EIPC01 10/24/05 1:42 PM Page 23 24 CHAPTER 1 drought-resistant, hard-leaved scrub of low-growing woody plants. Annual plants are also common in chaparral regions dur- ing the winter and early spring, when rainfall is more abundant. Chaparral is subject to periodic fires; many plants produce seeds that will only germinate after fire while others can quickly resprout because of food reserves in their fire-resistant roots. Deserts (see Plate 1.5, between pp. XX and XX) are found in areas that experi- ence extreme water shortage: rainfall is usually less than about 25 cm year−1, is usually very unpredictable and is considerably less than potential evaporation. The desert biome spans a very wide range of temperatures, from hot deserts, such as the Sahara, to very cold deserts, such as the Gobi in Mongolia. In their most extreme form, the hot deserts are too arid to bear any vegetation; they are as bare as the cold deserts of Antarctica. Where there is sufficient rainfall to allow plants to grow in arid deserts, its timing is always unpredictable. Desert vegetation falls into two sharply contrasted patterns of behavior. Many species have an opportunistic lifestyle, stimulated into germination by the unpredictable rains. They grow fast and complete their life history by starting to set new seed after a few weeks. These are the species that can occasionally make a desert bloom. A different pattern of behavior is to be long-lived with sluggish physiological processes. Cacti and other succulents, and small shrubby species with small, thick and often hairy leaves, can close their stomata (pores through which gas exchange takes place) and tolerate long periods of physiological inactivity. The relative poverty of animal life in arid deserts reflects the low productiv- ity of the vegetation and the indigestibility of much of it. Tropical rainforest (see Plate 1.6, between pp. XX and XX) is the most productive of the earth’s biomes – a result of the coincidence of high solar radiation received through- out the year and regular and reliable rainfall. The productivity is achieved, overwhelmingly, high in the dense forest canopy of evergreen foliage. It is dark at ground level except where fallen trees create gaps. Often, many tree seedlings and saplings remain in a suppressed state from year to year and only leap into action if a gap forms in the canopy above them. Apart from the trees, the vegetation is largely composed of plant forms that reach up into the canopy vicariously; they either climb and then scramble in the tree canopy (vines and lianas, including many species of fig) or grow as epiphytes, rooted on the damp upper branches. Most species of both animals and plants in tropical rain forest are active throughout the year, though the plants may flower and ripen fruit in sequence. Dramatically high species richness is the norm for tropical rainforest, and communities rarely if ever become dom- inated by one or a few species. The diversity of rainforest trees provides for a corresponding diversity of resources for herbivores, and so on up the food chain. Erwin (1982) estimated that there are 18,000 species of beetle in 1 ha of Panamanian rainforest (compared with only 24,000 in the whole of the United States and Canada!). All of these biomes are terrestrial. Aquatic ecologists could also come up with a set of biomes, although the tra- dition has largely been a terrestrial one. We might distinguish springs, rivers, ponds, lakes, estuaries, coastal zones, coral reefs and deep oceans, among other distinctive kinds of aquatic com- munity. For present purposes, we recognize just two aquatic biomes, marine and freshwater. The oceans cover about 71% of the earth’s surface and reach depths of more than 10,000 m. They extend from regions where precipitation exceeds evapora- tion to regions where the opposite is true. There are massive move- ments within this body of water that prevent major differences in salt concentrations developing (the average concentration is about 3%). Two main factors influence the biological activity of the oceans. Photosynthetically active radiation is absorbed in its pas- sage through water, so photosynthesis is confined to the surface region. Mineral nutrients, especially nitrogen and phosphorus, are commonly so dilute that they limit the biomass that can develop. Shallow waters (e.g. coastal regions and estuaries) tend to have high biological activity because they receive mineral input from the land and less incident radiation is lost than in passage through deep waters. Intense biological activity also occurs where nutrient-rich waters from the ocean depths come to the surface; this accounts for the concentration of many of the world’s fisheries in Arctic and Antarctic waters. Freshwater biomes occur mainly on the route from land drainage to the sea. The chemical composition of the water varies enormously, depending on its source, its rate of flow and the inputs of organic matter from vegetation that is rooted in or around the aquatic environment. In water catchments where the rate of evaporation is high, salts leached from the land may accumulate and the concentrations may far exceed those present in the oceans; brine lakes or even salt pans may be formed in which little life is possible. Even in aquatic situations liquid water may be unavailable, as is the case in the polar regions. Differentiating between biomes allows only a very crude recognition of the sorts of differences and similarities that occur between communities of organisms. Within biomes there are both small- and large-scale patterns of variation in the structure of com- munities and in the organisms that inhabit them. Moreover, as we see next, what characterizes a biome is not necessarily the particular species that live there. 1.5.2 The ‘life form spectra’ of communities We pointed out earlier the crucial importance of geographic isolation in allowing populations to diverge under selection. The geographic distributions of species, genera, families and even higher taxonomic categories of plants and animals often reflect this geographic divergence. All species of lemurs, for example, are found on the island of Madagascar and nowhere else. Similarly, •••• desert tropical rainforest aquatic biomes? EIPC01 10/24/05 1:42 PM Page 24 THE EVOLUTIONARY BACKDROP 27 daily and seasonal cycles), directional (like the accumulation of a pollutant in a lake) or erratic (like fires, hailstorms and typhoons). Heterogeneity crops up again and again in later chapters – in part because of the challenges it poses to organisms in moving from patch to patch (Chapter 6), in part because of the variety of opportunities it provides for different species (Chapters 8 and 19), and in part because heterogeneity can alter communities by interrupting what would otherwise be a steady march to an equilibrium state (Chapters 10 and 19). 1.6.2 Pairs of species As we have already noted, the existence of one type of organism in an area immediately diversifies it for others. Over its lifetime, an organism may increase the diversity of its environment by con- tributing dung, urine, dead leaves and ultimately its dead body. During its life, its body may serve as a place in which other species find homes. Indeed, some of the most strongly developed matches between organisms and their environment are those in which one species has developed a dependence upon another. This is the case in many relationships between consumers and their foods. Whole syndromes of form, behavior and metabolism constrain the animal within its narrow food niche, and deny it access to what might otherwise appear suitable alternative foods. Similar tight matches are characteristic of the relationships between parasites and their hosts. The various interactions in which one species is consumed by another are the subject matter of Chapters 9–12. Where two species have evolved a mutual dependence, the fit may be even tighter. We examine such ‘mutualisms’ in detail in Chapter 13. The association of nitrogen-fixing bacteria with the roots of leguminous plants, and the often extremely precise rela- tionships between insect pollinators and their flowers, are two good examples. When a population has been exposed to variations in the phys- ical factors of the environment, for example a short growing season or a high risk of frost or drought, a once-and-for-all toler- ance may ultimately evolve. The physical factor cannot itself change or evolve as a result of the evolution of the organisms. By contrast, when members of two species interact, the change in each produces alterations in the life of the other, and each may generate selective forces that direct the evolution of the other. In such a coevolutionary process the interaction between two species may continually escalate. What we then see in nature may be pairs of species that have driven each other into ever narrowing ruts of specialization – an ever closer match. •••• N um be r of s pe ci es ( % ) CI 0 HF 20 30 40 M (a) 10 CI 0 HF 20 30 40 M (b) 10 N um be r of s pe ci es ( % ) CI 0 HF 20 30 40 M (c) 10 CI 0 HF 20 30 40 M (d) 10 Figure 1.20 The percentages of forest mammals in various locomotory and feeding habitat categories in communities in: (a) Malaya, all forested areas (161 species), (b) Panama dry forest (70 species), (c) Australia, Cape York forest (50 species), and (d) Zaire, Irangi forest (96 species). C, carnivores; HF, herbivores and fructivores; I, insectivores; M, mixed feeders; ( ) aerial; ( ) arboreal; ( ) scansorial; ( ) small ground mammals. (After Andrews et al., 1979.) EIPC01 10/24/05 1:42 PM Page 27 28 CHAPTER 1 1.6.3 Coexistence of similar species While it is no surprise that species with rather different roles coex- ist within the same community, it is also generally the case that communities support a variety of species performing apparently rather similar roles. The Antarctic seals are an example. It is thought that the ancestral seals evolved in the northern hemisphere, where they are present as Miocene fossils, but one group of seals moved south into warmer waters and probably colonized the Antarctic in the Late Miocene or Early Pliocene (about 5 million years ago). When they entered the Antarctic, the Southern Ocean was probably rich in food and free from major predators, as it is today. It was within this environment that the group appears to have undergone radiative evolution (Figure 1.21). For ex- ample, the Weddell seal feeds primarily on fish and has unspe- cialized dentition; the crab-eater seal feeds almost exclusively on krill and its teeth are suited to filtering these from the sea water; the Ross seal has small, sharp teeth and feeds mainly on pelagic squid; and the leopard seal has large, cusped, grasping teeth and feeds on a wide variety of foods, including other seals and, in some seasons, penguins. Do these species compete with one another? Do competing species need to be different if they are to coexist? If so, how dif- ferent do they need to be: is there some limit to their similarity? Do species like the seals interact with one another at the present time, or has evolution in the past led to the absence of such interactions in contemporary communities? We return to these questions about coexisting, similar species in Chapter 8. Even at this stage, though, we may note that coexisting species, even when apparently very similar, commonly differ in subtle ways – not simply in their morphology or physiology but also in their responses to their environment and the role they play within the community of which they are part. The ‘ecological niches’ of such species are said to be differentiated from one another. The concept of the ecological niche is itself explained in the next two chapters. •••• (c) (a) (d) (b) Figure 1.21 Antarctic seals, similar species that coexist: (a) the Weddell seal, Leptonychotes weddellii (© Imageshop – zefa visual media uk ltd/Alamy), (b) the crab-eater seal Lobodon carcinophagus (© Bryan & Cherry Alexander Photography/Alamy), (c) the Ross seal, Omatophoca rossii (© Chris Sattlberger/Science Photo Library), and (d) the leopard seal, Hydrurga leptonyx (© Kevin Schafer/Alamy). EIPC01 10/24/05 1:42 PM Page 28 THE EVOLUTIONARY BACKDROP 29 Summary ‘Nothing in biology makes sense, except in the light of evolution’. We try in this chapter to illustrate the processes by which the properties of different sorts of species make their life possible in particular environments. We explain what is meant by evolutionary adaptation and by the theory of evolution by natural selection, an ecological theory first elaborated by Charles Darwin in 1859. Through natural selection, organisms come to match their environments by being ‘the fittest available’ or ‘the fittest yet’: they are not ‘the best imaginable’. Adaptive variation within species can occur at a range of levels: all represent a balance between local adaptation and hybridization. Ecotypes are genetically determined variants between populations within a species that reflect local matches between the organisms and their environments. Genetic poly- morphism is the occurrence together in the same habitat of two or more distinct forms. Dramatic examples of local specialization have been driven by manmade ecological forces, especially those of environmental pollution. We describe the process of speciation by which two or more new species are formed from one original species and explain what we mean by a ‘species’, especially a biospecies. Islands provide arguably the most favorable environment for populations to diverge into distinct species. Species live where they do for reasons that are often accidents of history. We illustrate this by examining island patterns, the move- ments of land masses over geological time, climatic changes especially during the Pleistocene ice ages (and we compare this with predicted changes consequent on current global warming) and the concepts of convergent and parallel evolution. The various terrestrial biomes of the earth are reviewed and their aquatic equivalents touched on briefly. Raunkiaer’s concept of life form spectra, in particular, emphasizes that ecological communities may be fundamentally very similar even when taxonomically quite distinct. All communities comprise a diversity of species: a diversity of matches to the local environment. Environmental heterogeneity, interactions between predators and prey, parasites and hosts and mutualists, and the coexistence of similar species all contribute to this. •••• EIPC01 10/24/05 1:42 PM Page 29 32 CHAPTER 2 2.3 Responses of individuals to temperature 2.3.1 What do we mean by ‘extreme’? It seems natural to describe certain environmental conditions as ‘extreme’, ‘harsh’, ‘benign’ or ‘stressful’. It may seem obvious when conditions are ‘extreme’: the midday heat of a desert, the cold of an Antarctic winter, the salinity of the Great Salt Lake. But this only means that these conditions are extreme for us, given our particular physiological characteristics and tolerances. To a cactus there is nothing extreme about the desert condi- tions in which cacti have evolved; nor are the icy fastnesses of Antarctica an extreme environment for penguins (Wharton, 2002). It is too easy and dangerous for the ecologist to assume that all other organisms sense the environment in the way we do. Rather, the ecologist should try to gain a worm’s-eye or plant’s-eye view of the environment: to see the world as others see it. Emotive words like harsh and benign, even relat- ivities such as hot and cold, should be used by ecologists only with care. •••• Ranunculus glacialis Oxyria digyna Geum reptans Pinus cembra Picea abies Betula pendula Larix decidua Picea abies Larix decidua Leucojum vernum Betula pendula Fagus sylvatica Taxus baccata Abies alba Prunus laurocerasus Quercus ilex Olea europaea Quercus pubescens Citrus limonum Te m pe ra tu re ( °C ) 25 20 15 10 5 Salinity (%) 0 10 20 30 405 15 25 35 45 2600 2500 2500 1900 1900 1900 1900 900 900 600 600 600 550 530 250 240 240 240 80 (m) (a) (b)Temperature (°C) 5 10 15 20 25 30 100% mortality 50% mortality Zero mortality Temperature pH (c) Fo od a va ila ble Figure 2.2 (a) A niche in one dimension. The range of temperatures at which a variety of plant species from the European Alps can achieve net photosynthesis of low intensities of radiation (70 W m−2). (After Pisek et al., 1973.) (b) A niche in two dimensions for the sand shrimp (Crangon septemspinosa) showing the fate of egg-bearing females in aerated water at a range of temperatures and salinities. (After Haefner, 1970.) (c) A diagrammatic niche in three dimensions for an aquatic organism showing a volume defined by the temperature, pH and availability of food. EIPC02 10/24/05 1:44 PM Page 32 CONDITIONS 33 2.3.2 Metabolism, growth, development and size Individuals respond to temperature essentially in the manner shown in Figure 2.1a: impaired function and ultimately death at the upper and lower extremes (discussed in Sec- tions 2.3.4 and 2.3.6), with a functional range between the extremes, within which there is an optimum. This is accounted for, in part, simply by changes in metabolic effectiveness. For each 10°C rise in temperature, for example, the rate of biological enzy- matic processes often roughly doubles, and thus appears as an exponential curve on a plot of rate against temperature (Figure 2.3). The increase is brought about because high temperature increases the speed of molecular movement and speeds up chemical reac- tions. The factor by which a reaction changes over a 10°C range is referred to as a Q10: a rough doubling means that Q10 ≈ 2. For an ecologist, however, effects on individual chemical reactions are likely to be less important than effects on rates of growth (increases in mass), on rates of development (progression through lifecycle stages) and on final body size, since, as we shall discuss much more fully in Chapter 4, these tend to drive the core ecological activities of survival, reproduction and movement. And when we plot rates of growth and development of whole organisms against temperature, there is quite com- monly an extended range over which there are, at most, only slight deviations from linearity (Figure 2.4). When the relationship between growth or development is effectively linear, the temperatures experienced by an organism can be summarized in a single very useful value, the number of ‘day- degrees’. For instance, Figure 2.4c shows that at 15°C (5.1°C above a development threshold of 9.9°C) the predatory mite, Amblyseius californicus, took 24.22 days to develop (i.e. the proportion of its total development achieved each day was 0.041 (= 1/24.22)), but it took only 8.18 days to develop at 25°C (15.1°C above the same threshold). At both temperatures, therefore, development required 123.5 day-degrees (or, more properly, ‘day-degrees above thresh- old’), i.e. 24.22 × 5.1 = 123.5, and 8.18 × 15.1 = 123.5. This is also the requirement for development in the mite at other temper- atures within the nonlethal range. Such organisms cannot be said to require a certain length of time for development. What they require is a combination of time and temperature, often referred to as ‘physiological time’. Together, the rates of growth and development determine the final size of an organism. For instance, for a given rate of growth, a faster rate of devel- opment will lead to smaller final size. Hence, if the responses of growth and development to variations in temperature are not the same, temperature will also affect final size. In fact, development usually increases more rapidly with temperature than does growth, such that, for a very wide range of organisms, final size tends to decrease with rearing temperature: the ‘temperature–size rule’ (see Atkinson et al., 2003). An example for single-celled protists (72 data sets from marine, brackish and freshwater habitats) is shown in Figure 2.5: for each 1°C increase in temperature, final cell volume decreased by roughly 2.5%. These effects of temperature on growth, development and size may be of practical rather than simply scientific importance. Increasingly, ecologists are called upon to predict. We may wish to know what the consequences would be, say, of a 2°C rise in temperature resulting from global warming (see Section 2.9.2). Or we may wish to understand the role of temperature in sea- sonal, interannual and geographic variations in the productivity of, for example, marine ecosystems (Blackford et al., 2004). We cannot afford to assume exponential relationships with temper- ature if they are really linear, nor to ignore the effects of changes in organism size on their role in ecological communities. Motivated, perhaps, by this need to be able to extrapolate from the known to the unknown, and also simply by a wish to discover fundamental organiz- ing principles governing the world •••• exponential effects of temperature on metabolic reactions effectively linear effects on rates of growth and development Temperature (°C) 5 10 15 20 25 30 O xy ge n co ns um pt io n (µ l O 2 g– 1 h– 1 ) 600 500 400 300 200 100 Figure 2.3 The rate of oxygen consumption of the Colorado beetle (Leptinotarsa decemineata), which doubles for every 10°C rise in temperature up to 20°C, but increases less fast at higher temperatures. (After Marzusch, 1952.) day-degree concept temperature–size rule ‘universal temperature dependence’? EIPC02 10/24/05 1:44 PM Page 33 •• 34 CHAPTER 2 around us, there have been attempts to uncover universal rules of temperature dependence, for metabolism itself and for develop- ment rates, linking all organisms by scaling such dependences with aspects of body size (Gillooly et al., 2001, 2002). Others have suggested that such generalizations may be oversimplified, stress- ing for example that characteristics of whole organisms, like growth and development rates, are determined not only by the temperature dependence of individual chemical reactions, but also by those of the availability of resources, their rate of diffusion from the environment to metabolizing tissues, and so on (Rombough, 2003; Clarke, 2004). It may be that there is room for coexistence between broad-sweep generalizations at the grand scale and the more complex relationships at the level of individual species that these generalizations subsume. 2.3.3 Ectotherms and endotherms Many organisms have a body temperature that differs little, if at all, from their environment. A parasitic worm in the gut of a mammal, a fungal mycelium in the soil and a sponge in the sea acquire the temperature of the medium in which they live. Terrestrial organisms, exposed to the sun and the air, are differ- ent because they may acquire heat directly by absorbing solar radi- ation or be cooled by the latent heat of evaporation of water (typical •• G ro w th r at e (µ m d ay –1 ) –0.2 4 1.0 Temperature (°C) 0.8 246 8 10 12 14 16 18 20 22 (a) 0.6 0.4 0.2 0.0 D ev el op m en ta l r at e 0 5 0.25 Temperature (°C) 0.2 0.15 0.1 0.05 3510 20 30 (c) 15 25 y = 0.0081x – 0.05 R 2 = 0.6838 D ev el op m en ta l r at e 0.08 18 0.2 Temperature (°C) 0.18 0.16 2820 22 24 26 (b) 0.14 0.12 0.1 y = 0.0124x – 0.1384 R 2 = 0.9753 y = 0.072x – 0.32 R 2 = 0.64 Figure 2.4 Effectively linear relationships between rates of growth and development and temperature. (a) Growth of the protist Strombidinopsis multiauris. (After Montagnes et al., 2003.) (b) Egg development in the beetle Oulema duftschmidi. (After Severini et al., 2003.) (c) Egg to adult development in the mite Amblyseius californicus. (After Hart et al., 2002.) The vertical scales in (b) and (c) represent the proportion of total development achieved in 1 day at the temperature concerned. (D iff er en ce fr om V 15 )/ V 15 –0.8 –20 1.2 Temperature (°C – 15) 20–10 0 10 0.8 0.4 0 –0.4 Figure 2.5 The temperature–size rule (final size decreases with increasing temperature) illustrated in protists (65 data sets combined). The horizontal scale measures temperature as a deviation from 15°C. The vertical scale measures standardized size: the difference between the cell volume observed and the cell volume at 15°C, divided by cell volume at 15°C. The slope of the mean regression line, which must pass through the point (0,0), was −0.025 (SE, 0.004); the cell volume decreased by 2.5% for every 1°C rise in rearing temperature. (After Atkinson et al., 2003.) EIPC02 10/24/05 1:44 PM Page 34 CONDITIONS 37 ‘chilling injury’. The fruits of the banana blacken and rot after exposure to chilling temperatures and many tropical rainforest species are sensitive to chilling. The nature of the injury is obscure, although it seems to be associated with the breakdown of membrane permeability and the leakage of specific ions such as calcium (Minorsky, 1985). Temperatures below 0°C can have lethal physical and chem- ical consequences even though ice may not be formed. Water may ‘supercool’ to temperatures at least as low as −40°C, remaining in an unstable liquid form in which its physical properties change in ways that are bound to be biologically significant: its viscosity increases, its diffusion rate decreases and its degree of ionization of water decreases. In fact, ice seldom forms in an organism until the temperature has fallen several degrees below 0°C. Body fluids remain in a supercooled state until ice forms suddenly around particles that act as nuclei. The concentration of solutes in the remaining liquid phase rises as a consequence. It is very rare for ice to form within cells and it is then inevitably lethal, but the freezing of extracellular water is one of the factors that prevents ice forming within the cells themselves (Wharton, 2002), since water is withdrawn from the cell, and solutes in the cytoplasm (and vacuoles) become more concentrated. The effects of freez- ing are therefore mainly osmoregulatory: the water balance of the cells is upset and cell membranes are destabilized. The effects are essentially similar to those of drought and salinity. Organisms have at least two differ- ent metabolic strategies that allow survival through the low temperatures of winter. A ‘freeze-avoiding’ strategy uses low-molecular-weight polyhydric alcohols (polyols, such as glycerol) that depress both the freezing and the supercooling point and also ‘thermal hysteresis’ proteins that prevent ice nuclei from forming (Figure 2.8a, b). A contrasting ‘freeze-tolerant’ strategy, which also involves the formation of polyols, encour- ages the formation of extracellular ice, but protects the cell membranes from damage when water is withdrawn from the cells (Storey, 1990). The tolerances of organisms to low temperatures are not fixed but are preconditioned by the experience of tem- peratures in their recent past. This process is called acclimation when it occurs in the laboratory and acclimatization when it occurs naturally. Acclimatization may start as the weather becomes colder in the fall, stimulating the conversion of almost the entire glycogen reserve of animals into polyols (Figure 2.8c), but this can be an energetically costly affair: about 16% of the carbohydrate reserve may be consumed in the conversion of the glycogen reserves to polyols. The exposure of an individual for several days to a relatively low tem- perature can shift its whole temperature response downwards along the tem- perature scale. Similarly, exposure to a high temperature can shift the temperature response upwards. Antarctic springtails (tiny arthropods), for instance, when taken from ‘summer’ temperat- ures in the field (around 5°C in the Antarctic) and subjected to a range of acclimation temperatures, responded to temperatures in the range +2°C to −2°C (indicative of winter) by showing a marked drop in the temperature at which they froze (Figure 2.9); but at lower acclimation temperatures still (−5°C, −7°C), they showed no such drop because the temperatures were themselves too low for the physiological processes required to make the acclimation response. Acclimatization aside, individuals commonly vary in their temperature response depending on the stage of development they have reached. Probably the most extreme form of this is when an organism has a dormant stage in its life cycle. Dormant stages are typically dehydrated, metabolically slow and tolerant of extremes of temperature. 2.3.5 Genetic variation and the evolution of cold tolerance Even within species there are often differences in temperature response between populations from different locations, and these differences have frequently been found to be the result of genetic differences rather than being attributable solely to acclimatization. Powerful evidence that cold tolerance varies between geographic races of a species comes from a study of the cactus, Opuntia fragilis. Cacti are generally species of hot dry habitats, but O. fragilis extends as far north as 56°N and at one site the lowest extreme minimum temperature recorded was −49.4°C. Twenty populations were sampled from diverse localities in northern USA and Canada, and were tested for freezing tolerance and ability to acclimate to cold. Individuals from the most freeze-tolerant population (from Manitoba) tolerated −49°C in laboratory tests and acclimated by 19.9°C, whereas plants from a population in the more equable climate of Hornby Island, British Columbia, tolerated only −19°C and acclimated by only 12.1°C (Loik & Nobel, 1993). There are also striking cases where the geographic range of a crop species has been extended into colder regions by plant breeders. Programs of deliberate selection applied to corn (Zea mays) have expanded the area of the USA over which the crop can be profitably grown. From the 1920s to the 1940s, the pro- duction of corn in Iowa and Illinois increased by around 24%, whereas in the colder state of Wisconsin it increased by 54%. If deliberate selection can change the tolerance and distribu- tion of a domesticated plant we should expect natural selection to have done the same thing in nature. To test this, the plant Umbilicus rupestris, which lives in mild maritime areas of Great Britain, was deliberately grown outside its normal range (Wood- ward, 1990). A population of plants and seeds was taken from a donor population in the mild-wintered habitat of Cardiff in the west and introduced in a cooler environment at an altitude of •••• freeze-avoidance and freeze-tolerance acclimation and acclimatization EIPC02 10/24/05 1:44 PM Page 37 •• •• 38 CHAPTER 2 Te m pe ra tu re ( °C ) –40 –20 0 20 (b) DecOctSep Nov AprMarFebJan G ly ce ro l c on ce nt ra tio n (µ m ol g –1 ) 0 1000 2000 3000 (a) DecOctSep Nov AprMarFebJan G ly co ge n co nc en tr at io n (µ m ol g –1 ) 0 400 800 1200 (c) DecOctSep Nov AprMarFebJan Month Figure 2.8 (a) Changes in the glycerol concentration per gram wet mass of the freeze-avoiding larvae of the goldenrod gall moth, Epiblema scudderiana. (b) The daily temperature maxima and minima (above) and whole larvae supercooling points (below) over the same period. (c) Changes in glycogen concentration over the same period. (After Rickards et al., 1987.) EIPC02 10/24/05 1:44 PM Page 38 •• CONDITIONS 39 157 m in Sussex in the south. After 8 years, the temperature response of seeds from the donor and the introduced populations had diverged quite strikingly (Figure 2.10a), and subfreezing temperatures that kill in Cardiff (−12°C) were then tolerated by 50% of the Sussex population (Figure 2.10b). This suggests that past climatic changes, for example ice ages, will have changed the temperature tolerance of species as well as forcing their migration. •• –6 –10 –14 –22 S up er co ol in g po in t ( °C ) Exposure temperature (°C) 1 –20 5 –3 –7 –8 –12 –18 –16 –5–13 Figure 2.9 Acclimation to low temperatures. Samples of the Antarctic springtail Cryptopygus antarcticus were taken from field sites in the summer (c. 5°C) on a number of days and their supercooling point (at which they froze) was determined either immediately (
) or after a period of acclimation (
) at the temperatures shown. The supercooling points of the controls themselves varied because of temperature variations from day to day, but acclimation at temperatures in the range +2 to −2°C (indicative of winter) led to a drop in the supercooling point, whereas no such drop was observed at higher temperatures (indicative of summer) or lower temperatures (too low for a physiological acclimation response). Bars are standard errors. (After Worland & Convey, 2001.) G er m in at io n (% ) 2216 0 6 40 80 10 Temperature (°C) (a) 2 1 S ur vi va l ( % ) –14–8 0 40 80 –4 Minimum temperature (°C) (b) 2 –12 1 Figure 2.10 Changes in the behavior of populations of the plant Umbilicus rupestris, established for a period of 8 years in a cool environment in Sussex from a donor population in a mild-wintered area in South Wales (Cardiff, UK). (a) Temperature responses of seed germination: (1) responses of samples from the donor population (Cardiff ) in 1978, and (2) responses from the Sussex population in 1987. (b) The low-temperature survival of the donor population at Cardiff, 1978 (1) and of the established population in Sussex, 1987 (2). (After Woodward, 1990.) EIPC02 10/24/05 1:44 PM Page 39 •••• 42 CHAPTER 2 –2 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 Year 2000 Niño 3.4 region (threshold − 0°C) 2 1 0 –1 S ea s ur fa ce te m pe ra tu re a no m al ie s 3 (a) Figure 2.11 (a) The El Niño–Southern Oscillation (ENSO) from 1950 to 2000 as measured by sea surface temperature anomalies (differences from the mean) in the equatorial mid-Pacific. The El Niño events (> 0.4°C above the mean) are shown in dark color, and the La Niña events (> 0.4°C below the mean) are shown in pale color. (Image from http://www.cgd.ucar.edu/cas/catalog/ climind/Nino_3_3.4_indices.html.) (b) Maps of examples of El Niño (November 1997) and La Niña (February 1999) events in terms of sea height above average levels. Warmer seas are higher; for example, a sea height 15–20 cm below average equates to a temperature anomaly of approximately 2–3°C. (Image from http://topex-www.jpl.nasa.gov/science/images/el-nino-la-nina.jpg.) (For color, see Plate 2.1, between pp. 000 and 000.) (b) EIPC02 10/24/05 1:44 PM Page 42 •••• CONDITIONS 43 –4 1860 1880 1900 1920 1940 1960 1980 6 (c) Year (L n – S n) 2000 2 4 0 –2 Figure 2.11 (continued) (c) The North Atlantic Oscillation (NAO) from 1864 to 2003 as measured by the normalized sea-level pressure difference (Ln − Sn) between Lisbon, Portugal and Reykjavik, Iceland. (Image from http://www.cgd.ucar.edu/~jhurrell/ nao.stat.winter.html#winter.) (d) Typical winter conditions when the NAO index is positive or negative. Conditions that are more than usually warm, cold, dry or wet are indicated. (Image from http://www.ldeo.columbia.edu/NAO/.) (For color, see Plate 2.2, between pp. 000 and 000.) (d)(i) (d)(ii) EIPC02 10/24/05 1:44 PM Page 43 44 CHAPTER 2 isotherm (Figure 2.14a; an isotherm is a line on a map joining places that experience the same temperature – in this case a January mean of 4.5°C). However, we need to be very careful how we inter- pret such relationships: they can be extremely valuable in predicting where we might and might not find a particular species; they may suggest that some feature related to temperature is import- ant in the life of the organisms; but they do not prove that tem- perature causes the limits to a species’ distribution. The literature relevant to this and many other correlations between temperature and distribution patterns is reviewed by Hengeveld (1990), who also describes a more subtle graphical procedure. The minimum temperature of the coldest month and the maximum temperature of the hottest month are estimated for many places within and outside the range of a species. Each location is then plotted on a graph of maximum against minimum temperature, and a line is drawn that optimally discriminates between the presence and absence records (Figure 2.14b). This line is then used to define the geographic margin of the species distributions (Figure 2.14c). This may have powerful predictive value, but it still tells us nothing about the underlying forces that cause the distribution patterns. One reason why we need to be cautious about reading too much into correlations of species distributions with maps of tem- perature is that the temperatures measured for constructing isotherms for a map are only rarely those that the organisms expe- rience. In nature an organism may choose to lie in the sun or hide •••• lo g( ab un da nc e ag e 3 in 1 00 0s ) 4.5 –5 8.0 NAO index 7.5 7.0 6.5 6.0 5.5 5.0 6–4 –3 –2 –1 0 1 2 3 4 5 lo g( ab un da nc e ag e 3 in 1 00 0s ) 4.5 50 8.0 Length of 5-month-old cod (mm) 7.5 7.0 6.5 6.0 5.5 5.0 10060 70 80 90 T em pe ra tu re ( °C ) 2.5 –5 5 NAO index 4.5 4 3.5 3 6–4 –3 –2 –1 0 1 2 3 4 5 Le ng th o f 5 -m on th -o ld c od ( m m ) 50 2.5 100 Temperature (°C) 90 80 70 60 53 3.5 4 4.5 (a) (d) (b) (c) Figure 2.12 (a) The abundance of 3-year-old cod, Gadus morhua, in the Barents Sea is positively correlated with the value of the North Atlantic Oscillation (NAO) index for that year. The mechanism underlying this correlation is suggested in (b–d). (b) Annual mean temperature increases with the NAO index. (c) The length of 5-month-old cod increases with annual mean temperature. (d) The abundance of cod at age 3 increases with their length at 5 months. (After Ottersen et al., 2001.) EIPC02 10/24/05 1:44 PM Page 44 CONDITIONS 47 of only 50%. The organisms most obviously affected by humid- ity in their distribution are those ‘terrestrial’ animals that are actually, in terms of the way they control their water balance, ‘aquatic’. Amphibians, terrestrial isopods, nematodes, earthworms and molluscs are all, at least in their active stages, confined to microenvironments where the relative humidity is at or very close to 100%. The major group of animals to escape such confinement are the terrestrial arthropods, especially insects. Even here though, the evaporative loss of water often confines their activities to habitats (e.g. woodlands) or times of day (e.g. dusk) when relative humidity is relatively high. 2.5 pH of soil and water The pH of soil in terrestrial environments or of water in aquatic ones is a condition that can exert a powerful influence on the dis- tribution and abundance of organisms. The protoplasm of the root cells of most vascular plants is damaged as a direct result of toxic concentrations of H+ or OH− ions in soils below pH 3 or above pH 9, respectively. Further, indirect effects occur because soil pH influences the availability of nutrients and/or the concentration of toxins (Figure 2.17). Increased acidity (low pH) may act in three ways: (i) directly, by upsetting osmoregulation, enzyme activity or gaseous exchange across respiratory surfaces; (ii) indirectly, by increasing the con- centration of toxic heavy metals, particularly aluminum (Al3+) but also manganese (Mn2+) and iron (Fe3+), which are essential plant nutrients at higher pHs; and (iii) indirectly, by reducing the qual- ity and range of food sources available to animals (e.g. fungal growth is reduced at low pH in streams (Hildrew et al., 1984) and the aquatic flora is often absent or less diverse). Tolerance limits for pH vary amongst plant species, but only a minority are able to grow and reproduce at a pH below about 4.5. In alkaline soils, iron (Fe3+) and phosphate (PO4 3+), and certain trace elements such as manganese (Mn2+), are fixed in relatively insoluble compounds, and plants may then suffer because there is too little rather than too much of them. For example, calcifuge plants (those characteristic of acid soils) commonly show symp- toms of iron deficiency when they are transplanted to more alka- line soils. In general, however, soils and waters with a pH above 7 tend to be hospitable to many more species than those that are more acid. Chalk and limestone grasslands carry a much richer flora (and associated fauna) than acid grasslands and the situation is similar for animals inhabiting streams, ponds and lakes. Some prokaryotes, especially the Archaebacteria, can tolerate and even grow best in environments with a pH far outside the range tolerated by eukaryotes. Such environments are rare, but occur in volcanic lakes and geothermal springs where they are •••• 1.0 0 S ur vi va l r at e fu nc tio n Experiment period (days) 100 200 0.5 0 100 2000 6°C 12°C S. malma S. leucomaenis Figure 2.16 Changing temperature reverses the outcome of competition. At low temperature (6°C) on the left, the salmonid fish Salvelinus malma outsurvives cohabiting S. leucomaenis, whereas at 12°C, on the right, S. leucomaenis drives S. malma to extinction. Both species are quite capable, alone, of living at either temperature. (After Taniguchi & Nakano, 2000.) 964 5 pH 3 87 Mo Fe and Mn Cu and Zn K Ca and Mg P and B N and S mobilization Al H+ and OH– toxicity Fgiure 2.17 The toxicity of H+ and OH− to plants, and the availability to them of minerals (indicated by the widths of the bands) is influenced by soil pH. (After Larcher, 1980.) EIPC02 10/24/05 1:44 PM Page 47 48 CHAPTER 2 dominated by sulfur-oxidizing bacteria whose pH optima lie between 2 and 4 and which cannot grow at neutrality (Stolp, 1988). Thiobacillus ferroxidans occurs in the waste from industrial metal- leaching processes and tolerates pH 1; T. thiooxidans cannot only tolerate but can grow at pH 0. Towards the other end of the pH range are the alkaline environments of soda lakes with pH values of 9–11, which are inhabited by cyanobacteria such as Anabaenopsis arnoldii and Spirulina platensis; Plectonema nostocorum can grow at pH 13. 2.6 Salinity For terrestrial plants, the concentration of salts in the soil water offers osmotic resistance to water uptake. The most extreme saline conditions occur in arid zones where the predominant movement of soil water is towards the surface and cystalline salt accumu- lates. This occurs especially when crops have been grown in arid regions under irrigation; salt pans then develop and the land is lost to agriculture. The main effect of salinity is to create the same kind of osmoregulatory problems as drought and freezing and the problems are countered in much the same ways. For example, many of the higher plants that live in saline environ- ments (halophytes) accumulate electrolytes in their vacuoles, but maintain a low concentration in the cytoplasm and organelles (Robinson et al., 1983). Such plants maintain high osmotic pres- sures and so remain turgid, and are protected from the damaging action of the accumulated electrolytes by polyols and membrane protectants. Freshwater environments present a set of specialized environ- mental conditions because water tends to move into organisms from the environment and this needs to be resisted. In marine habitats, the majority of organisms are isotonic to their environ- ment so that there is no net flow of water, but there are many that are hypotonic so that water flows out from the organism to the environment, putting them in a similar position to terrestrial organisms. Thus, for many aquatic organisms the regulation of body fluid concentration is a vital and sometimes an energetically expensive process. The salinity of an aquatic environment can have an important influence on distribution and abundance, especially in places like estuaries where there is a particularly sharp gradi- ent between truly marine and freshwater habitats. The freshwater shrimps Palaemonetes pugio and P. vulgaris, for example, co-occur in estuaries on the eastern coat of the USA at a wide range of salinities, but the former seems to be more tolerant of lower salinities than the latter, occupying some habitats from which the latter is absent. Figure 2.18 shows the mechanism likely to be underlying this (Rowe, 2002). Over the low salinity range (though not at the effectively lethal lowest salin- ity) metabolic expenditure was significantly lower in P. pugio. P. vulgaris requires far more energy simply to maintain itself, putting it at a severe disadvantage in competition with P. pugio even when it is able to sustain such expenditure. 2.6.1 Conditions at the boundary between the sea and land Salinity has important effects on the distribution of organisms in intertidal areas but it does so through interactions with other conditions – notably exposure to the air and the nature of the substrate. •••• S ta nd ar d m et ab ol ic e xp en di tu re ( J da y– 1 ) 33 32 31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 Salinity (ppt) 0 1 2 3 4 5 6 7 353025201510 Overall mean, P. vulgaris (24.85) Overall mean, P. pugio (22.91) P. pugio P. vulgaris Figure 2.18 Standard metabolic expenditure (estimated through minimum oxygen consumption) in two species of shrimp, Palaemonetes pugio and P. vulgaris, at a range of salinities. There was significant mortality of both species over the experimental period at 0.5 ppt (parts per thousand), especially in P. vulgaris (75% compared with 25%). (After Rowe, 2002.) EIPC02 10/24/05 1:44 PM Page 48 CONDITIONS 49 Algae of all types have found suitable habitats permanently immersed in the sea, but permanently submerged higher plants are almost completely absent. This is a striking contrast with submerged freshwater habitats where a variety of flowering plants have a conspicuous role. The main reason seems to be that higher plants require a substrate in which their roots can find anchorage. Large marine algae, which are continuously sub- merged except at extremely low tides, largely take their place in marine communities. These do not have roots but attach themselves to rocks by specialized ‘holdfasts’. They are excluded from regions where the substrates are soft and holdfasts cannot ‘hold fast’. It is in such regions that the few truly marine flower- ing plants, for example sea grasses such as Zostera and Posidonia, form submerged communities that support complex animal communities. Most species of higher plants that root in seawater have leaves and shoots that are exposed to the atmosphere for a large part of the tidal cycle, such as mangroves, species of the grass genus Spartina and extreme halo- phytes such as species of Salicornia that have aerial shoots but whose roots are exposed to the full salinity of seawater. Where there is a stable substrate in which plants can root, communities of flowering plants may extend right through the intertidal zone in a continuum extending from those continuously immersed in full-strength seawater (like the sea grasses) through to totally non- saline conditions. Salt marshes, in particular, encompass a range of salt concentrations running from full-strength seawater down to totally nonsaline conditions. Higher plants are absent from intertidal rocky sea shores except where pockets of soft substrate may have formed in crevices. Instead, such habitats are dominated by the algae, which give way to lichens at and above the high tide level where the exposure to desiccation is highest. The plants and animals that live on rocky sea shores are influenced by environmental condi- tions in a very profound and often particularly obvious way by the extent to which they tolerate exposure to the aerial environ- ment and the forces of waves and storms. This expresses itself in the zonation of the organisms, with different species at different heights up the shore (Figure 2.19). The extent of the intertidal zone depends on the height of tides and the slope of the shore. Away from the shore, the tidal rise and fall are rarely greater than 1 m, but closer to shore, the shape of the land mass can funnel the ebb and flow of the water to produce extraordinary spring tidal ranges of, for example, nearly 20 m in the Bay of Fundy (between Nova Scotia and New Brunswick, Canada). In contrast, the shores of the Mediterranean Sea •••• Figure 2.19 A general zonation scheme for the seashore determined by relative lengths of exposure to the air and to the action of waves. (After Raffaelli & Hawkins, 1996.) Land Sea Supra littora l zone Up per lim it o f la min atio n s eaw eed s Upp er li mit of b arna cles Uppe r lim it of periw inkle sna ils Supralittoral fringe Midlittoral zone Infralittoral zone Infralittoral fringe Littoral zone algae and higher plants zonation EIPC02 10/24/05 1:44 PM Page 49 •• 52 CHAPTER 2 recorded in the succession of diatom species accumulated in lake sediments (Flower et al., 1994). Figure 2.22, for example, shows how diatom species composition has changed in Lough Maam, Ireland – far from major industrial sites. The percentage of vari- ous diatom species at different depths reflects the flora present at various times in the past (four species are illustrated). The age of layers of sediment can be determined by the radioactive decay of lead-210 (and other elements). We know the pH tolerance of the diatom species from their present distribution and this can be used to reconstruct what the pH of the lake has been in the past. Note how the waters acidified since about 1900. The diatoms Fragilaria virescens and Brachysira vitrea have declined markedly dur- ing this period while the acid-tolerant Cymbella perpusilla and Frustulia rhomboides increased after 1900. 2.9 Global change In Chapter 1 we discussed some of the ways in which global environments have changed over the long timescales involved in continental drift and the shorter timescales of the repeated ice ages. Over these timescales some organisms have failed to accommodate to the changes and have become extinct, others have migrated so that they continue to experience the same conditions but in a different place, and it is probable that others have changed their nature (evolved) and tolerated some of the changes. We now turn to consider global changes that are occur- ring in our own lifetimes – consequences of our own activities – and that are predicted, in most scenarios, to bring about profound changes in the ecology of the planet. 2.9.1 Industrial gases and the greenhouse effect A major element of the Industrial Revolution was the switch from the use of sustainable fuels to the use of coal (and later, oil) as a source of power. Between the middle of the 19th and the middle of the 20th century the burning of fossil fuels, together with extensive deforestation, added about 9 × 1010 tonnes of carbon dioxide (CO2) to the atmosphere and even more has been added since. The concentration of CO2 in the atmosphere before the Industrial Revolution (measured in gas trapped in ice cores) was about 280 ppm, a fairly typical interglacial ‘peak’ (Figure 2.23), but this had risen to around 370 ppm by around the turn of the millennium and is still rising (see Figure 18.22). Solar radiation incident on the earth’s atmosphere is in part reflected, in part absorbed, and part is transmitted through to the earth’s surface, which absorbs and is warmed by it. Some of this absorbed energy is radiated back to the atmosphere where atmo- spheric gases, mainly water vapor and CO2 absorb about 70% of it. It is this trapped reradiated energy that heats the atmosphere in what is called the ‘greenhouse effect’. The greenhouse effect was of course part of the normal environment before the Industrial Revolution and carried responsibility for some of the environmental warmth before industrial activity started to enhance it. At that time, atmospheric water vapor was respons- ible for the greater portion of the greenhouse effect. •• 25 10 10 25 100 250 100 25 10 500 1000 100 250 1000 2000 3000 4000 1000 100250 500 500 2000 10 00 500 Figure 2.21 An example of long-distance environmental pollution. The distribution in Great Britain of fallout of radioactive caesium (Bq m−2) from the Chernobyl nuclear accident in the Soviet Union in 1986. The map shows the persistence of the pollutant on acid upland soils where it is recycled through soils, plants and animals. Sheep in the upland areas contained more caesium-137 (137Cs) in 1987 and 1988 (after recycling) than in 1986. 137Cs has a half-life of 30 years! On typical lowland soils it is more quickly immobilized and does not persist in the food chains. (After NERC, 1990.) EIPC02 10/24/05 1:44 PM Page 52 •• CONDITIONS 53 In addition to the enhancement of greenhouse effects by increased CO2, other trace gases have increased markedly in the atmosphere, particularly methane (CH4) (Figure 2.24a; and compare this with the his- torical record in Figure 2.23), nitrous oxide (N2O) and the chlorofluorocarbons (CFCs, e.g. trichlorofluoromethane (CCl3F) and dichlorodifluoromethane (CCl2F2)). Together, these and other gases contribute almost as much to enhancing the green- house effect as does the rise in CO2 (Figure 2.24b). The increase in CH4 is not all explained but probably has a microbial origin in intensive agriculture on anaerobic soils (especially increased rice production) and in the digestive process of ruminants (a cow pro- duces approximately 40 litres of CH4 each day); around 70% of its production is anthropogenic (Khalil, 1999). The effect of the CFCs from refrigerants, aerosol propellants and so on is poten- tially great, but international agreements at least appear to have halted further rises in their concentrations (Khalil, 1999). It should be possible to draw up a balance sheet that shows how the CO2 produced by human activities translates into the changes in concentration in the atmosphere. Human activities •• Percent 0 3010 B ra ch ys ira v itr ea 20 Date A.D. 1988 1969 1940 1903 5.2 5.4 5.6 5.8 6.0 pH 0 10 20 30 40 0 3010 F ra gi la ria v ire sc en s 200 10 F ru st ul ia r ho m bo id es S ed im en t d ep th ( cm ) 0 40 5 10 15 30 25 20 0 20 35 10 C ym be lla p er pu si lla Figure 2.22 The history of the diatom flora of an Irish lake (Lough Maam, County Donegal) can be traced by taking cores from the sediment at the bottom of the lake. The percentage of various diatom species at different depths reflects the flora present at various times in the past (four species are illustrated). The age of the layers of sediment can be determined by the radioactive decay of lead-210 (and other elements). We know the pH tolerance of the diatom species from their present distribution and this can be used to reconstruct what the pH of the lake has been in the past. Note how the waters have been acidified since about 1900. The diatoms Fragilaria virescens and Brachysira vitrea have declined markedly during this period, while the acid-tolerant Cymbella perpusilla and Frustulia rhomboides have increased. (After Flower et al., 1994.) CO2 – but not only CO2 C O 2 (p pm ) 0100,000300,000400,000 200 240 280 200,000 Age BP (years) C H 4 (p pb ) 400 600 700 500 Figure 2.23 Concentrations of CO2 and methane (CH4) in gas trapped in ice cores from Vostok, Antarctica deposited over the past 420,000 years. Estimated temperatures are very strongly correlated with these. Thus, transitions between glacial and warm epochs occurred around 335,000, 245,000, 135,000 and 18,000 years ago. BP, before present; ppb, parts per billion; ppm, parts per million. (After Petit et al., 1999; Stauffer, 2000.) EIPC02 10/24/05 1:44 PM Page 53 54 CHAPTER 2 release 5.1–7.5 × 109 metric tons of carbon to the atmosphere each year. But the increase in atmospheric CO2 (2.9 × 109 metric tons) accounts for only 60% of this, a percentage that has remained remarkably constant for 40 years (Hansen et al., 1999). The oceans absorb CO2 from the atmosphere, and it is estimated that they may absorb 1.8–2.5 × 109 metric tons of the carbon released by human activities. Recent analyses also indicate that terrestrial vegetation has been ‘fertilized’ by the increased atmospheric CO2, so that a considerable amount of extra carbon has been locked up in vegetation biomass (Kicklighter et al., 1999). This softening of the blow by the oceans and terrestrial vegetation notwith- standing, however, atmospheric CO2 and the greenhouse effect are increasing. We return to the question of global carbon budgets in Section 18.4.6. 2.9.2 Global warming We started this chapter discussing temperature, moved through a number of other environmental conditions to pollutants, and now return to temperature because of the effects of those pollu- tants on global temperatures. It appears that the present air temperature at the land surface is 0.6 ± 0.2°C warmer than in preindustrial times (Figure 2.25), and temperatures are predicted to continue to rise by a further 1.4–5.8°C by 2100 (IPCC, 2001). Such changes will probably lead to a melting of the ice caps, a consequent rising of sea level and large changes in the pattern of global climates and the distribution of species. Predictions of the extent of global warming resulting from the enhanced green- house effect come from two sources: (i) predictions based on sophisticated computer models (‘general circulation models’) that simulate the world’s climate; and (ii) trends detected in mea- sured data sets, including the width of tree rings, sea-level records and measures of the rate of retreat of glaciers. Not surprisingly, different global circulation models differ in their pre- dictions of the rise in global tempera- ture that will result from predicted increases in CO2. However, most model predictions vary only from 2.3 to 5.2°C (most of the variation is accounted for by the way in which the effects of cloud cover are modeled), and a projected rise of 3–4°C in the next 100 years seems a reasonable value from which to make projections of ecological effects (Figure 2.26). But temperature regimes are, of course, only part of the set of conditions that determine which organisms live where. Unfortunately, we can place much less faith in computer projec- tions of rainfall and evaporation because it is very hard to build good models of cloud behavior into a general model of climate. If we consider only temperature as a relevant variable, we would project a 3°C rise in temperature giving London (UK) the climate of Lisbon (Portugal) (with an appropriate vegetation of olives, vines, Bougainvillea and semiarid scrub). But with more reliable rain it would be nearly subtropical, and with a little less it might qualify for the status of an arid zone! •••• C on ce nt ra te d C H 4 (p pb ) 200019601920 800 1900 1400 1600 1800 1940 Year (a) 1000 1200 1980 C al cu la te d te m pe ra tu re c ha ng e (° C ) 0.0 CO2 0.3 0.4 0.5 Trace gas (b) 0.1 0.2 CFCsCH4 N2O Figure 2.24 (a) Concentration of methane (CH4) in the atmosphere through the 20th century. (b) Estimates of global warming over the period 1850–1990 caused by CO2 and other major greenhouse gases. (After Khalil, 1999.) a 3–4°C rise in the next 100 years EIPC02 10/24/05 1:44 PM Page 54 CONDITIONS 57 modern concept – Hutchinson’s n-dimensional hypervolume – also distinguishes fundamental and realized niches. Temperature is discussed in detail as a typical, and perhaps the most important, condition. Individuals respond to temperature with impaired function and ultimately death at upper and lower extremes, with a functional range between the extremes, within which there is an optimum, although these responses may be subject to evolutionary adaptation and to more immediate acclimatization. The rates of biological enzymatic processes often increase exponentially with temperature (often Q10 ≈ 2), but for rates of growth and development there are often only slight deviations from linearity: the basis for the day-degree concept. Because development usually increases more rapidly with temperature than does growth, final size tends to decrease with rearing temperature. Attempts to uncover universal rules of temperature dependence remain a matter of controversy. We explain the differences between endotherms and ecto- therms but also the similarities between them, ultimately, in their responses to a range of temperatures. We examine variations in temperature on and within the sur- face of the earth with a variety of causes: latitudinal, altitudinal, continental, seasonal, diurnal and microclimatic effects, and, in soil and water, the effects of depth. Increasingly, the importance of medium-term temporal patterns have become apparent. Notable amongst these are the El Niño–Southern Oscillation (ENSO) and the North Atlantic Oscillation (NAO). There are very many examples of plant and animal distribu- tions that are strikingly correlated with some aspect of environ- mental temperature but these do not prove that temperature directly causes the limits to a species’ distribution. The tem- peratures measured are only rarely those that the organisms experience. For many species, distributions are accounted for not so much by average temperatures as by occasional extremes; and the effects of temperature may be determined largely by the responses of other community members or by interactions with other conditions. A range of other environmental conditions are also discussed: the pH of soil and water, salinity, conditions at the boundary between sea and land, and the physical forces of winds, waves and currents. Hazards, disasters and catastrophes are distinguished. A number of environmental conditions are becoming increas- ingly important due to the accumulation of toxic by-products of human activities. A striking example is the creation of ‘acid rain’. Another is the effect of industrial gases on the greenhouse effect and consequent effects on global warming. A projected rise of 3–4°C in the next 100 years seems a reasonable value from which to make projections of ecological effects, though global warming is not evenly distributed over the surface of the earth. This rate is 50–100 times faster than postglacial warming. We must expect latitudinal and altitudinal changes to species’ distributions and widespread extinctions of floras and faunas. •••• EIPC02 10/24/05 1:44 PM Page 57 •• 3.1 Introduction According to Tilman (1982), all things consumed by an organism are resources for it. But consumed does not simply mean ‘eaten’. Bees and squirrels do not eat holes, but a hole that is occupied is no longer available to another bee or squirrel, just as an atom of nitrogen, a sip of nectar or a mouthful of acorn are no longer available to other consumers. Similarly, females that have already mated may be unavailable to other mates. All these things have been consumed in the sense that the stock or supply has been reduced. Thus, resources are entities required by an organ- ism, the quantities of which can be reduced by the activity of the organism. Green plants photosynthesize and obtain both energy and matter for growth and reproduction from inorganic materials. Their resources are solar radiation, carbon dioxide (CO2), water and mineral nutrients. ‘Chemosynthetic’ organisms, such as many of the Archaebacteria, obtain energy by oxidizing methane, ammonium ions, hydrogen sulfide or ferrous iron; they live in environments such as hot springs and deep sea vents and use resources that were much more abundant during early phases of life on earth. All other organisms use as their food resource the bodies of other organisms. In each case, what has been consumed is no longer available to another consumer. The rabbit eaten by an eagle is no longer available to another eagle. The quantum of solar radiation absorbed and photosynthesized by a leaf is no longer available to another leaf. This has an important consequence: organ- isms may compete with each other to capture a share of a limited resource – a topic that will occupy us in Chapter 5. A large part of ecology is about the assembly of inorganic resources by green plants and the reassembly of these packages at each successive stage in a web of consumer–resource inter- actions. In this chapter we start with the resources of plants and focus especially on those most important in photosynthesis: radiation and CO2. Together, plant resources fuel the growth of individual plants, which, collectively, determine the primary productivity of whole areas of land (or volumes of water): the rate, per unit area, at which plants produce biomass. Patterns of prim- ary productivity are examined in Chapter 17. Relatively little space in this chapter is given to food as a resource for animals, simply because a series of later chapters (9–12) is devoted to the ecology of predators, grazers, parasites and saprotrophs (the consumers and decomposers of dead organisms). This chapter then closes where the previous chapter began: with the ecological niche, adding resource dimensions to the condition dimensions we have met already. 3.2 Radiation Solar radiation is the only source of energy that can be used in metabolic activities by green plants. It comes to the plant as a flux of radiation from the sun, either directly having been diffused to a greater or lesser extent by the atmosphere, or after being reflected or transmitted by other objects. The direct fraction is highest at low latitudes (Figure 3.1). Moreover, for much of the year in temperate climates, and for the whole of the year in arid climates, the leaf canopy in terrestrial communities does not cover the land surface, so that most of the incident radiation falls on bare branches or on bare ground. When a plant intercepts radiant energy it may be reflected (with its wavelength unchanged), transmitted (after some wavebands have been filtered out) or absorbed. Part of the fraction that is absorbed may raise the plant’s temperature and be reradiated at much longer wavelengths; in terrestrial plants, part may contribute latent heat of evaporation of water and so power the transpiration what are resources? organisms may compete for resources the fate of radiation Chapter 3 Resources EIPC03 10/24/05 1:47 PM Page 58 RESOURCES 59 stream. A small part may reach the chloroplasts and drive the process of photosynthesis (Figure 3.2). Radiant energy is converted during photosynthesis into energy-rich chem- ical compounds of carbon, which will subsequently be broken down in re- spiration (either by the plant itself or by organisms that consume it). But unless the radiation is cap- tured and chemically fixed at the instant it falls on the leaf, it is irretrievably lost for photosynthesis. Radiant energy that has been fixed in photosynthesis passes just once through the world. This is in complete contrast to an atom of nitrogen or carbon or a molecule of water that may cycle repeatedly through endless generations of organisms. Solar radiation is a resource con- tinuum: a spectrum of different wave- lengths. But the photosynthetic apparatus is able to gain access to energy in only a restricted band of this spectrum. All green plants depend on chlorophyll and other pigments for the photosynthetic fixation of carbon, and these pigments fix radiation in a waveband between roughly 400 and 700 nm. This is the band of ‘photosynthetically active radiation’ (PAR). It corresponds broadly with the range of the spectrum visible to the human eye that we call ‘light’. About 56% of the radiation incident on the earth’s surface lies outside the PAR range and is thus unavailable as a resource for green plants. In other organisms there are pigments, for example bacterio- chlorophyll in bacteria, that operate in photosynthesis outside the PAR range of green plants. 3.2.1 Variations in the intensity and quality of radiation A major reason why plants seldom achieve their intrinsic photosynthetic capacity is that the intensity of radiation varies continually (Figure 3.3). Plant morphology and physiology that are optimal for photosynthesis at one intensity of radiation will usually be inappropriate at another. In terrestrial habitats, leaves live in a radiation regime that varies throughout the day and the year, and they live in an environment of other leaves that modifies the quantity and quality of radiation received. As with all resources, the supply of radiation can vary both systematically (diurnal, annual) and unsystematically. Moreover, it is not the case simply that the inten- sity of radiation is a greater or lesser proportion of a maximum value at which photosynthesis would be most productive. At high intensities, photoinhibition of photosynthesis may occur (Long et al., 1994), such that the rate of fixation of carbon decreases with increasing radiation intensity. High intensities of radiation may also lead to dangerous overheating of plants. Radiation is an essential resource for plants, but they can have too much as well as too little. Annual and diurnal rhythms are systematic variations in solar radiation (Figure 3.3a, b). The green plant expe- riences periods of famine and glut in its radiation resource every 24 h (except near the poles) and seasons of famine and glut every year (except in the tropics). In aquatic habitats, an additional •• radiant energy must be captured or is lost forever photosynthetically active radiation 2.1 2.1 1.68 1.68 1.68 1.68 1.26 0.84 2.1 2.1 1.68 1.68 1.68 1.68 2.1 1.68 1.26 0.84 0.84 1.26 2.1 1.68 1.26 1.68 2.12.1 2.1 1.68 0.84 Figure 3.1 Global map of the solar radiation absorbed annually in the earth– atmosphere system: from data obtained with a radiometer on the Nimbus 3 meteorological satellite. The units are J cm−2 min−1. (After Raushke et al., 1973.) photoinhibition at high intensities systematic variations in supply EIPC03 10/24/05 1:47 PM Page 59 62 CHAPTER 3 rhythm in the intensity of radiation – they are shed when they are least useful. In consequence, an evergreen leaf of an under- story species may experience a further systematic change, because the seasonal cycle of leaf production of overstory species deter- mines what radiation remains to penetrate to the understory. The daily movement of leaves in many species also reflects the changing intensity and direction of incident radiation. Less systematic variations in the radiation environment of a leaf are caused by the nature and position of neighboring leaves. Each canopy, each plant and each leaf, by intercepting radiation, creates a resource- depletion zone (RDZ) – a moving band of shadow over other leaves of the same plant, or of others. Deep in a canopy, shadows become less well defined because much of the radiation loses its original direction by diffusion and reflection. Submerged vegetation in aquatic habitats is likely to have a much less sys- tematic shading effect, simply because it is moved around by the flow of the water in which it lives, though vegeta- tion floating on the surface, especially of ponds or lake, inevitably has a profound and largely unvary- ing effect on the radiation regime beneath it. Phytoplankton cells nearer the surface, too, shade the cells beneath them, such that the reduction of intensity with depth is greater, the greater the phytoplankton density. Figure 3.4, for example, shows the decline in light penetration, measured at a set depth in a laboratory system, as a population of the unicellular green alga, Chlorella vulgaris, built up over a 12-day period (Huisman, 1999). The composition of radiation that has passed through leaves in a canopy, or through a body of water, is also altered. It may be less useful photo- synthetically because the PAR component has been reduced – though such reductions may also, of course, prevent photo- inhibition and overheating. Figure 3.5 shows an example for the variation with depth in a freshwater habitat. The major differences amongst ter- restrial species in their reaction to sys- tematic variations in the intensity of radiation are those that have evolved between ‘sun species’ and ‘shade species’. In general, plant species that are characteristic of shaded habitats use radiation at low intensities more efficiently than sun species, but the reverse is true at high intensities (Figure 3.6). Part of the difference between them lies in the physiology of the leaves, but the mor- phology of the plants also influences the efficiency with which radiation is captured. The leaves of sun plants are commonly exposed at acute angles to the midday sun (Poulson & DeLucia, 1993). This spreads an incident beam of radiation over a larger leaf area, and effectively reduces its intensity. An intensity of radiation that is superoptimal for photosynthesis when it strikes a leaf at 90° may therefore be optimal for a leaf inclined at an acute angle. The leaves of sun plants are often superimposed into •••• shade: a resource- depletion zone attenuation with depth, and plankton density, in aquatic habitats variations in quality as well as quantity sun and shade species 20 60 40 20 0 15 10 5 0 Time (days) 24201612840 P op ul at io n de ns ity ( ce lls m l– 1 ) × 1 0– 6 Li gh t p en et ra tio n (µ m ol p ho to ns m –2 s –1 ) Figure 3.4 As population density (
) of the unicellular green alga, Chlorella vulgaris, increased in laboratory culture, this increased density reduced the penetration of light (7; its intensity at a set depth). Bars are standard deviations; they are omitted when they are smaller than the symbols. (After Huisman, 1999.) Q ua nt um ir ra di an ce (1 01 5 qu an ta m –2 s –1 n m –1 ) 5000 4000 3000 2000 1000 0 Wavelength (nm) 750700650600550500450400 0 m 5 m (×25) 3 m Figure 3.5 Changing spectral distribution of radiation with depth in Lake Burley Griffin, Australia. Note that photosynthetically active radiation lies broadly within the range 400–700 nm. (After Kirk, 1994.) EIPC03 10/24/05 1:47 PM Page 62 RESOURCES 63 a multilayered canopy. In bright sunshine even the shaded leaves in lower layers may have positive rates of net photosynthesis. Shade plants commonly have leaves held near to the horizontal and in a single-layered canopy. In contrast to these ‘strategic’ dif- ferences, it may also happen that as a plant grows, its leaves develop differently as a ‘tactical’ response to the radiation environment in which it developed. This often leads to the formation of ‘sun leaves’ and ‘shade leaves’ within the canopy of a single plant. Sun leaves are typically smaller, thicker, have more cells per unit area, denser veins, more densely packed chloroplasts and a greater dry weight per unit area of leaf. These tactical maneuvers, then, tend to occur not at the level of the whole plant, but at the level of the individual leaf or even its parts. Nevertheless, they take time. To form sun or shade leaves as a tactical response, the plant, its bud or the developing leaf must sense the leaf ’s environment and respond by growing a leaf with an appropriate structure. For exam- ple, it is impossible for the plant to change its form fast enough to track the changes in intensity of radiation between a cloudy and a clear day. It can, however, change its rate of photosyn- thesis extremely rapidly, reacting even to the passing of a fleck of sunlight. The rate at which a leaf photosynthesizes also depends on the demands that are made on it by other vigorously growing parts. Photosynthesis may be reduced, even though conditions are otherwise ideal, if there is no demanding call on its products. In aquatic habitats, much of the variation between species is accounted for by differences in photosynthetic pigments, which contribute significantly to the precise wave- lengths of radiation that can be utilized (Kirk, 1994). Of the three types of pigment – chlorophylls, carotenoids and biliproteins – all photosynthetic plants contain the first two, but many algae also contain biliproteins; and within the chlorophylls, all higher plants have chlorophyll a and b, but many algae have only chlorophyll a and some have chlorophyll a and c. Examples of the absorp- tion spectra of a number of pigments, the related contrasting absorption spectra of a number of groups of aquatic plants, and the related distributional differences (with depth) between a number of groups of aquatic plants are illustrated in Figure 3.7. A detailed assessment of the evidence for direct links between pigments, performance and distribution is given by Kirk (1994). 3.2.2 Net photosynthesis The rate of photosynthesis is a gross measure of the rate at which a plant captures radiant energy and fixes it in organic carbon compounds. However, it is often more important to consider, and very much easier to measure, the net gain. Net photosynthesis is the increase (or decrease) in dry matter that results from the difference between gross photosynthesis and the losses due to respiration and the death of plant parts (Figure 3.8). Net photosynthesis is negative in darkness, when respiration exceeds photosynthesis, and increases with the intensity of PAR. The compensation point is the intensity of PAR at which the gain from gross photosyn- thesis exactly balances the respiratory and other losses. The leaves of shade species tend to respire at lower rates than those of sun species. Thus, when both are growing in the shade the net photo- synthesis of shade species is greater than that of sun species. There is nearly a 100-fold variation in the photosynthetic capacity of leaves (Mooney & Gulmon, 1979). This is the rate of photosynthesis when incident radiation is saturating, temperature is optimal, relative humidity is high, and CO2 and oxygen concentrations are normal. When the leaves of different species are compared under these ideal conditions, the ones with the highest photosynthetic capacity are generally those from environments where nutrients, water and radiation are seldom limiting (at least during the growing season). These include many agricultural crops and their weeds. Species from resource-poor environments (e.g. shade plants, desert perennials, heathland species) usually have low photosynthetic capacity – even when abundant resources are provided. Such pat- terns can be understood by noting that photosynthetic capacity, like all capacity, must be ‘built’; and the investment in building •••• C O 2 up ta ke ( m g C O 2 dm –2 h –1 ) 0 0 30 40 50 Radiation intensity (100 J m–2 s–1) 987654321 20 10 C4 C3 Shade herbs Shade mosses, planktonic algae Beech Sun herbs Wheat Corn Sorghum 10 Figure 3.6 The response of photosynthesis to light intensity in various plants at optimal temperatures and with a natural supply of CO2. Note that corn and sorghum are C4 plants and the remainder are C3 (the terms are explained in Sections 3.3.1 and 3.3.2). (After Larcher, 1980, and other sources.) sun and shade leaves the compensation point photosynthetic capacity pigment variation in aquatic species EIPC03 10/24/05 1:47 PM Page 63 •••• 64 CHAPTER 3 A bs or ba nc e Macrophyte 750 0.0 1.0 Wavelength (nm) 600500400 0.8 0.6 0.4 0.2 700650550450 (e) A bs or ba nc e (d) 700 0.0 0.9 300 Wavelength (nm) 600500400 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 R-phycocyanin (b) A bs or ba nc e 700550450 0.00 0.75 1.00 450 Wavelength (nm) 650600 0.25 449 500 0.50 628 Chlorophyll c2 A bs or ba nc e β-carotene 550 0.0 2.0 2.5 400 Wavelength (nm) 500 1.0 450 1.5 0.5 (c) A bs or ba nc e Green algae(f) (a rb itr ar y un its ) 750 Wavelength (nm) 600500400 700650550450 (a) A bs or ba nc e 700550450 0.0 1.0 1.5 500 650600 0.5 Chlorophyll b Chlorophyll a Chlorophyll a and b Wavelength (nm) Figure 3.7 (a) Absorption spectra of chlorophylls a and b. (b) Absorption spectrum of chlorophyll c2. (c) Absorption spectrum of β-carotene. (d) Absorption spectrum of the biliprotein, R-phycocyanin. (e) Absorption spectrum of a piece of leaf of the freshwater macrophyte, Vallisneria spiralis, from Lake Ginnindera, Australia. (f ) Absorption spectrum of the planktonic alga Chlorella pyrenoidos (green). EIPC03 10/24/05 1:47 PM Page 64 RESOURCES 67 Overall, therefore, despite receiving only one-seventh of the PFD of sun plants, shade plants reduced the differential in the amount absorbed to one-quarter, and reduced the differential in their daily rate of carbon gain to only a half. Shade plants successfully counterbalanced their reduced photosynthetic capa- city at the leaf level with enhanced light-harvesting ability at the whole plant level. The sun plants can be seen as striking a compromise between maximizing whole plant photosynthesis on the one hand while avoiding photoinhibition and overheating of individual leaves on the other. 3.2.4 Photosynthesis or water conservation? Strategic and tactical solutions In fact, in terrestrial habitats especially, it is not sensible to consider radiation as a resource independently of water. Intercepted radiation does not result in photosynthesis unless there is CO2 available, and the prime route of entry of CO2 is through open stomata. But if the stomata are open to the air, water will evaporate through them. If water is lost faster than it can be gained, the leaf (and the plant) will sooner or later wilt and eventually die. But in most terres- trial communities, water is, at least sometimes, in short supply. Should a plant conserve water at the expense of present photo- synthesis, or maximize photosynthesis at the risk of running out of water? Once again, we meet the problem of whether the optimal solution involves a strict strategy or the ability to make tactical responses. There are good examples of both solutions and also compromises. Perhaps the most obvious strategy that plants may adopt is to have a short life and high photosynthetic activity during periods when water is abundant, but remain dormant as seeds during the rest of the year, neither photosynthesizing nor transpiring (e.g. many desert annuals, annual weeds and most annual crop plants). •••• Table 3.1 (a) Observed differences in the shoots and leaves of sun and shade plants of the shrub Heteromeles arbutifolia. Standard deviations are given in parentheses; the significance of differences are given following analysis of variance. (b) Consequent whole plant properties of sun and shade plants. (After Valladares & Pearcy, 1998.) (a) Sun Shade P Internode distance (cm) 1.08 (0.06) 1.65 (0.02) < 0.05 Leaf angle (degrees) 71.3 (16.3) 5.3 (4.3) < 0.01 Leaf surface area (cm2) 10.1 (0.3) 21.4 (0.8) < 0.01 Leaf blade thickness (mm) 462.5 (10.9) 292.4 (9.5) < 0.01 Photosynthetic capacity, area basis (mmol CO2 m −2 s−1) 14.1 (2.0) 9.0 (1.7) < 0.01 Photosynthetic capacity, mass basis (mmol CO2 kg −1 s−1) 60.8 (10.1) 58.1 (11.2) NS Chlorophyll content, area basis (mg m−2) 280.5 (15.3) 226.7 (14.0) < 0.01 Chlorophyll content, mass basis (mg g−1) 1.23 (0.04) 1.49 (0.03) < 0.05 Leaf nitrogen content, area basis (g m−2) 1.97 (0.25) 1.71 (0.21) < 0.05 Leaf nitrogen content, mass basis (% dry weight) 0.91 (0.31) 0.96 (0.30) NS (b) Sun plants Shade plants Summer Winter Summer Winter EP 0.55 a 0.80b 0.88b 0.54a ED 0.33 a 0.38a, b 0.41b 0.43b Fraction self-shaded 0.22a 0.42b 0.47b 0.11a EA, direct PFD 0.28 a 0.44b 0.55c 0.53c LARc (cm 2 g−1) 7.1a 11.7b 20.5c 19.7c EP, projection efficiency; ED, display efficiency; EA, absorption efficiency; LARe, effective leaf area ratio; NS, not significant. Letter codes indicate groups that differed significantly in analyses of variance (P < 0.05). stomatal opening short active interludes in a dormant life EIPC03 10/24/05 1:47 PM Page 67 68 CHAPTER 3 Second, plants with long lives may produce leaves during periods when water is abundant and shed them during droughts (e.g. many species of Acacia). Some shrubs of the Israeli desert (e.g. Teucrium polium) bear finely divided, thin- cuticled leaves during the season when soil water is freely avail- able. These are then replaced by undivided, small, thick-cuticled leaves in more drought-prone seasons, which in turn fall and may leave only green spines or thorns (Orshan, 1963): a sequential polymorphism through the season, with each leaf morph being replaced in turn by a less photosynthetically active but more water- tight structure. Next, leaves may be produced that are long lived, transpire only slowly and tolerate a water deficit, but which are unable to photosynthesize rapidly even when water is abundant (e.g. evergreen desert shrubs). Structural features such as hairs, sunken stomata and the restriction of stomata to specialized areas on the lower surface of a leaf slow down water loss. But these same morphological features reduce the rate of entry of CO2. Waxy and hairy leaf surfaces may, however, reflect a greater proportion of radiation that is not in the PAR range and so keep the leaf temperature down and reduce water loss. Finally, some groups of plants have evolved particular physiologies: C4 and crassulacean acid metabolism (CAM). We consider these in more detail in Sections 3.3.1–3.3.3. Here, we simply note that plants with ‘nor- mal’ (i.e. C3) photosynthesis are wasteful of water compared with plants that possess the modified C4 and CAM physiologies. The water-use efficiency of C4 plants (the amount of carbon fixed per unit of water transpired) may be double that of C3 plants. The viability of alternative strat- egies to solve a common problem is nicely illustrated by the trees of seasonally dry tropical forests and woodlands (Eamus, 1999). These communities are found naturally in Africa, the Americas, Australia and India, and as a result of human interference elsewhere in Asia. But whereas, for example, the savannas of Africa and India are dominated by deciduous species, and the Llanos of South America are dominated by evergreens, the savannas of Australia are occu- pied by roughly equal numbers of species from four groups (Figure 3.10a): evergreens (a full canopy all year), deciduous species (losing all leaves for at least 1 and usually 2–4 months each year), semideciduous species (losing around 50% or more of their leaves each year) and brevideciduous species (losing only about 20% of their leaves). At the ends of this continuum, the decidu- ous species avoid drought in the dry season (April–November in Australia) as a result of their vastly reduced rates of transpir- ation (Figure 3.10b), but the evergreens maintain a positive carbon balance throughout the year (Figure 3.10c), whereas the deciduous species make no net photosynthate at all for around 3 months. The major tactical control of the rates of both photosynthe- sis and water loss is through changes in stomatal ‘conductance’ that may occur rapidly during the course of a day and allow a very rapid response to immediate water shortages. Rhythms of stomatal opening and closure may ensure that the above-ground parts of the plant remain more or less watertight except during controlled periods of active photosynthesis. These rhythms may •••• P er ce nt ag e ca no py fu lln es s J 0 J 20 60 100 Month (a) 40 80 F M A M J J A S O N D P re da w n w at er p ot en tia l M P a) J –2.0 J 0.0 Month (b) –1.0 –0.5 F M A M J J A S O N D –1.5 A ss im ila tio n ra te ( µm ol m –2 s –1 ) J 0 J 2 12 16 Month (c) 10 14 F M A M J J A S O N D 4 6 8 Figure 3.10 (a) Percentage canopy fullness for deciduous (), semideciduous (), brevideciduous () and evergreen (
) trees Australian savannas throughout the year. (Note that the southern hemisphere dry season runs from around April to November.) (b) Susceptibility to drought as measured by increasingly negative values of ‘predawn water potential’ for deciduous () and evergreen (
) trees. (c) Net photosynthesis as measured by the carbon assimilation rate for deciduous () and evergreen (
) trees. (After Eamus, 1999.) leaf appearance and structure physiological strategies coexisting alternative strategies in Australian savannas EIPC03 10/24/05 1:47 PM Page 68 RESOURCES 69 be diurnal or may be quickly responsive to the plant’s internal water status. Stomatal movement may even be triggered directly by conditions at the leaf surface itself – the plant then responds to desiccating conditions at the very site, and at the same time, as the conditions are first sensed. 3.3 Carbon dioxide The CO2 used in photosynthesis is obtained almost entirely from the atmo- sphere, where its concentration has risen from approximately 280 µl l−1 in 1750 to about 370 µl l−1 today and is still increasing by 0.4–0.5% year−1 (see Figure 18.22). In a terrestrial community, the flux of CO2 at night is upwards, from the soil and vegetation to the atmosphere; on sunny days above a photosynthesizing canopy, there is a downward flux. Above a vegetation canopy, the air becomes rapidly mixed. However, the situation is quite different within and beneath canopies. Changes in CO2 con- centration in the air within a mixed deciduous forest in New England were measured at various heights above ground level during the year (Figure 3.11a) (Bazzaz & Williams, 1991). Highest concentrations, up to around 1800 µl l−1, were measured near the surface of the ground, tapering off to around 400 µl l−1 at 1 m above the ground. These high values near ground level were achieved in the summer when high temperatures allowed the rapid decomposition of litter and soil organic matter. At greater heights within the forest, the CO2 concentrations scarcely ever (even in winter) reached the value of 370 µl l−1 which is the atmospheric concentration of bulk air measured at the Mauna Loa laboratory in Hawaii (see Figure 18.22). During the winter months, concentrations remained virtually constant through the day and night at all heights. But in the summer, major diurnal cycles of concentration developed that reflected the interaction between the production of CO2 by decomposition and its con- sumption in photosynthesis (Figure 3.11b). That CO2 concentrations vary so widely within vegetation means that plants growing in different parts of a forest will experience quite different CO2 environments. Indeed the lower leaves on a forest shrub will usually experience higher CO2 concentrations than its upper leaves, and seedlings will live in environments richer in CO2 than mature trees. In aquatic environments, variations in CO2 concentration can be just as striking, especially when water mixing is limited, for example during the sum- mer ‘stratification’ of lakes, with layers of warm water towards the surface and colder layers beneath (Figure 3.12). •••• C O 2 co nc en tr at io ns ( µl l– 1 ) Dec 31Nov 11Apr 25Mar 6 440 Sep 22 Measurement date (a) 420 400 380 360 340 320 300 Jun 14 Aug 3 Time of day 0400 255 455 405 355 305 20001200 Jul 4 C O 2 co nc en tr at io ns ( µl l– 1 ) 0400 255 (b) 455 405 355 305 20001200 Nov 21 Figure 3.11 (a) CO2 concentrations in a mixed deciduous forest (Harvard Forest, Massachusetts, USA) at various times of year at five heights above ground: , 0.05 m; 4, 0.20 m; , 3.00 m; 7, 6.00 m;
, 12.00 m. Data from the Mauna Loa CO2 observatory (5) are given on the same axis for comparison. (b) CO2 concentrations for each hour of the day (averaged over 3–7-day periods) on November 21 and July 4. (After Bazzaz & Williams, 1991.) the rise in global levels variations beneath a canopy variations in aquatic habitats . . . EIPC03 10/24/05 1:47 PM Page 69 72 CHAPTER 3 and the clearing of forests. As Loladze (2002) points out, while consequential changes to global climate may be controversial in some quarters, marked increases in CO2 concentration itself are not. Plants now are experiencing around a 30% higher concentra- tion compared to the pre-industrial period – effectively instantan- eous on geological timescales; trees living now may experience a doubling in concentration over their lifetimes – effectively an instantaneous change on an evolutionary timescale; and high mixing rates in the atmosphere mean that these are changes that will affect all plants. There is also evidence of large- scale changes in atmospheric CO2 over much longer timescales. Carbon balance models suggest that during the Triassic, Jurassic and Cretaceous periods, atmospheric concen- trations of CO2 were four to eight times greater than at present, falling after the Cretaceous from between 1400 and 2800 µl l−1 to below 1000 µl l−1 in the Eocene, Miocene and Pliocene, and fluctuating between 180 and 280 µl l−1 during subsequent glacial and interglacial periods (Ehleringer & Monson, 1993). The declines in CO2 concentration in the atmosphere after the Cretaceous may have been the primary force that favored the evo- lution of plants with C4 physiology (Ehleringer et al., 1991), because at low concentrations of CO2, photorespiration places C3 plants at a particular disadvantage. The steady rise in CO2 since the Industrial Revolution is therefore a partial return to pre- Pleistocene conditions and C4 plants may begin to lose some of their advantage. •••• (a) 0.37 0.00 1.40 1.50 0.45 0.56 1.34 0.99 2.13 1.77 2.84 3.20 3.36 4.38 2.04 1.77 0.69 0.17 0.24 0.38 0.41 0.080.29 0.81 0.81 0.38 0.72 0.56 0.00 0.00 0.00 0.00 0.00 0.15 0.28 0.43 2.54 0.31 0.22 C 4 sp ec ie s (% ) 806535 0 20 1 2 50 Mean summer pan evaporation (inches per summer) (b) 4 3 r = 0.947 Figure 3.14 (a) The percentage of native C4 dicot species in various regions of North America. (b) The relationship between the percentage of native C4 species in 31 geographic regions of North America, and the mean summer (May–October) pan evaporation – a climatic indicator of plant/water balance. Regions for which appropriate climatic data were unavailable were excluded, together with south Florida, where the peculiar geography and climate may explain the aberrant composition of the flora. (After Stowe & Teeri, 1978.) changes in geological time EIPC03 10/24/05 1:47 PM Page 72 RESOURCES 73 When other resources are present at adequate levels, additional CO2 scarcely influences the rate of photo- synthesis of C4 plants but increases the rate of C3 plants. Indeed, artificially increasing the CO2 concen- tration in greenhouses is a commercial technique to increase crop (C3) yields. We might reasonably predict dramatic increases in the productivity of individual plants and of whole crops, forests and natural communities as atmospheric concentrations of CO2 continue to increase. In the 1990s alone, results from more than 2700 studies on free-air CO2 enrichment (FACE) experiments were published, and it is clear that, for example, doubling CO2 concentration generally stimulates photosynthesis and increases agricultural yield by an average of 41% (Loladze, 2002). However, there is also much evidence that the responses may be complicated (Bazzaz, 1990). For example, when six species of temperate forest tree were grown for 3 years in a CO2-enriched atmosphere in a glasshouse, they were generally larger than controls, but the CO2 enhancement of growth declined even within the relatively short timescale of the experiment (Bazzaz et al., 1993). Moreover, there is a general tendency for CO2 enrichment to change the composition of plants, and in particular to reduce nitrogen concentration in above-ground plant tissues – around 14% on average under CO2 enhancement (Cotrufo et al., 1998). This in turn may have indirect effects on plant–animal interac- tions, because insect herbivores may then eat 20–80% more foliage to maintain their nitrogen intake and fail to gain weight as fast (Figure 3.15). CO2 enhancement may also reduce concentrations in plants of other essential nutrients and micronutrients (Figure 3.16) (see Section 3.5), con- tributing in turn to ‘micronutrient malnutrition’, which diminishes the health and economy of more than one-half of the world’s human population (Loladze, 2002). 3.4 Water The volume of water that becomes incorporated in higher plants during growth is infinitesimal in comparison to the volume that flows through the plant in the transpiration stream. Nevertheless, water is a critical resource. Hydration is a necessary condition for metabolic reactions to proceed, and because no organism is completely watertight its water content needs continual replen- ishment. Most terrestrial animals drink free water and also gen- erate some from the metabolism of food and body materials; there are extreme cases in which animals of arid zones may obtain all their water from their food. 3.4.1 Roots as water foragers For most terrestrial plants, the main source of water is the soil and they gain access to it through a root system. We proceed here •••• M ea n la rv al w ei gh t ( m g) 3010 0 0 600 20 Larval age (days) 500 400 300 200 100 High CO2 Low CO2 Figure 3.15 Growth of larvae of the buckeye butterfly ( Junonia coenia) feeding on Plantago lanceolata that had been grown at ambient and elevated CO2 concentrations. (After Fajer, 1989.) what will be the consequences of current rises? CO2 and nitrogen and micronutrient composition N P K S Mg Fe Zn Mn CuCaC ha ng e in m ea n co nc en tr at io n (% ) –25.0 –15.0 5.0 Element –5.0Figure 3.16 Changes in the concentrations of nutrients in plant material grown at twice-ambient atmospheric CO2 concentrations, based on 25 studies on leaves of a variety of plants (colored bars) and five studies of wheat grains (gray bars). Black lines indicate the standard errors. (After Loladze, 2002.) EIPC03 10/24/05 1:47 PM Page 73 74 CHAPTER 3 (and in the next section on plant nutrient resources) on the basis of plants simply having ‘roots’. In fact, most plants do not have roots – they have mycorrhizae: associations of fungal and root tissue in which both partners are crucial to the resource-gathering properties of the whole. Mycorrhizae, and the respective roles of the plants and the fungi, are discussed in Chapter 13. It is not easy to see how roots evolved by the modification of any more primitive organ (Harper et al., 1991), yet the evolution of the root was almost certainly the most influential event that made an extensive land flora and fauna possible. Once roots had evolved they provided secure anchorage for structures the size of trees and a means for making intimate contact with mineral nutrients and water within the soil. Water enters the soil as rain or melting snow and forms a reservoir in the pores between soil particles. What happens to it then depends on the size of the pores, which may hold it by capillary forces against gravity. If the pores are wide, as in a sandy soil, much of the water will drain away until it reaches some impediment and accumulates as a rising watertable or finds its way into streams or rivers. The water held by soil pores against the force of gravity is called the ‘field capacity’ of the soil. This is the upper limit of the water that a freely drained soil will retain. There is a less clearly defined lower limit to the water that can be used in plant growth (Figure 3.17). This is determined by the ability of plants to extract water from the narrower soil pores, and is known as the ‘permanent wilting point’ – the soil water content at which plants wilt and are unable to recover. The permanent wilting point does not differ much between the plant species of mesic environments (i.e. with a moderate amount of water) or between species of crop plants, but many species native to arid regions can extract significantly more water from the soil. As a root withdraws water from the soil pores at its surface, it creates water-depletion zones around it. These determine gradi- ents of water potential between the interconnected soil pores. Water flows along the gradient into the depleted zones, supply- ing further water to the root. This simple process is made much more complex because the more the soil around the roots is depleted of water, the more resistance there is to water flow. As the root starts to withdraw water from the soil, the first water that it obtains is from the wider pores because they hold the water with weaker capillary forces. This leaves only the narrower, more tortuous water-filled paths through which flow can occur, and so the resistance to water flow increases. Thus, when the root draws water from the soil very rapidly, the resource depletion zone (RDZ; see Section 3.2.1) becomes sharply defined and water can move across it only slowly. For this reason, rapidly transpiring •••• 7640 5 pF1 2 3 –1000–10.0–0.001 Bars–0.1 0.00010.0010.11000 0.01 Pore size (µm) 100 10 Water unavailableAvailable water to many native species Water drains away freely Available water Rootlet diameters Root hairs Bacterial cells F ie ld c ap ac it y P er m an en t w ilt in g p o in t (s p ec ie s sp ec if ic ) Figure 3.17 The status of water in the soil, showing the relationship between three measures of water status: (i) pF, the logarithm of the height (cm) of the column of water that the soil would support; (ii) water status expressed as atmospheres or bars; (iii) the diameter of soil pores that remain water-filled. The size of water-filled pores may be compared in the figure with the sizes of rootlets, root hairs and bacterial cells. Note that for most species of crop plant the permanent wilting point is at approximately −15 bars (−1.5 × 106 Pa), but in many other species it reaches −80 bars (−8 ×106 Pa), depending on the osmotic potentials that the species can develop. field capacity and the permanent wilting point EIPC03 10/24/05 1:47 PM Page 74 •••• (a) Quercus alba Quercus ilicifolia Pinus rigida Vaccinium vacillans (b) Wood Bark Roots Leaves Flowers Fruits N P K Ca Mg S Fe Na Figure 3.19 (a) The relative concentration of various minerals in whole plants of four species in the Brookhaven Forest, New York. (b) The relative concentration of various minerals in different tissues of the white oak (Quercus alba) in the Brookhaven Forest. Note that the differences between species are much less than between the parts of a single species. (After Woodwell et al., 1975). D ep th ( m ) 3.0 0.5 (a) 2.5 2.0 1.5 1.0 0 Mc Pt Aps Ap Pt BdMcBg Sm Sand Sand Clay (b) Figure 3.20 (a) The root systems of plants in a typical short-grass prairie after a run of years with average rainfall (Hays, Kansas). Ap, Aristida purpurea; Aps, Ambrosia psilostachya; Bd, Buchloe dactyloides; Bg, Bouteloua gracilis; Mc, Malvastrum coccineum; Pt, Psoralia tenuiflora; Sm, Solidago mollis. (After Albertson, 1937; Weaver & Albertson, 1943.) (b) The root system developed by a plant of wheat grown through a sandy soil containing a layer of clay. Note the responsiveness of root development to the localized environment that it encounters. (Courtesy of J.V. Lake.) EIPC03 10/24/05 1:47 PM Page 77 78 CHAPTER 3 intimately branched, root systems. The less freely that water moves in the soil, the narrower will be the RDZs, and the more it will pay the plant to explore the soil intensively rather than extensively. The soil solution that flows through soil pores to the root surface has a biased mineral composition compared with what is potentially available. This is because different mineral ions are held by different forces in the soil. Ions such as nitrate, calcium and sodium may, in a fertile agricultural soil, be carried to the root surface faster than they are accumu- lated in the body of the plant. By contrast, the phosphate and potas- sium content of the soil solution will often fall far short of the plant’s requirements. Phosphate is bound on soil colloids by surfaces that bear calcium, aluminum and ferric ions, and the rate at which it can be extracted by plants then depends on the rate at which its concentration is replenished by release from the colloids. In dilute solutions, the diffusion coefficients of ions that are not absorbed, such as nitrate, are of the order of 10−5 cm2 s−l, and for cations such as calcium, magnesium, ammonium and potas- sium they are 10−7 cm2 s−1. For strongly absorbed anions such as phosphate, the coefficients are as low as 10−9 cm2 s−1. The diffusion rate is the main factor that determines the width of an RDZ. For resources like phosphate that have low diffusion coeffici- ents, the RDZs will be narrow (Figure 3.21); roots or root hairs will only tap common pools of resource (i.e. will compete) if they are very close together. It has been estimated that more than 90% of the phosphate absorbed by a root hair in a 4-day period will have come from the soil within 0.1 mm of its surface. Two roots will therefore only draw on the same phosphate resource in this period if they are less than 0.2 mm apart. A widely spaced, extensive root system tends to maximize access to nitrate, whilst a narrowly spaced, intensively branched root system tends to maximize access to phosphates (Nye & Tinker, 1977). Plants with different shapes of root system may therefore tolerate dif- ferent levels of soil mineral resources, and different species may deplete different mineral resources to different extents. This may be of great importance in allowing a variety of plant species to cohabit in the same area (coexistence of competitors is discussed in Chapters 8 and 19). 3.6 Oxygen Oxygen is a resource for both animals and plants. Only a few prokaryotes can do without it. Its diffusibility and solubility in water are very low and so it becomes limiting most quickly in aquatic and waterlogged environments. Its solubility in water also decreases rapidly with increasing temperature. When organic matter decomposes in an aquatic environment, microbial respira- tion makes a demand for oxygen and this ‘biological oxygen demand’ may constrain the types of higher animal that can persist. High biological oxygen demands are particularly characteristic of still waters into which leaf litter or organic pollutants are deposited and they become most acute during periods of high temperature. Because oxygen diffuses so slowly in water, aquatic animals must either maintain a continual flow of water over their respir- atory surfaces (e.g. the gills of fish), or have very large surface areas relative to body volume (e.g. many aquatic crustacea have large feathery appendages), or have specialized respiratory pigments or a slow respiration rate (e.g. the midge larvae that live in still and nutrient-rich waters), or continually return to the surface to breathe (e.g. whales, dolphins, turtles and newts). The roots of many higher plants fail to grow into water- logged soil, or die if the water table rises after they have penetrated deeply. These reactions may be direct responses to oxygen deficiency or responses to the accumulation of gases such as hydrogen sulfide, methane and ethylene, which are produced by microorganisms engaged in anaerobic decomposition. Even if roots do not die when starved of oxygen, they may cease to absorb min- eral nutrients so that the plants suffer from mineral deficiencies. 3.7 Organisms as food resources Autotrophic organisms (green plants and certain bacteria) assimilate inorganic resources into packages of organic •••• variations between nutrients in their freedom of movement Figure 3.21 Radioautograph of soil in which seedlings of mustard have been grown. The soil was supplied with radioactively labeled phosphate (32PO4 −) and the zones that have been depleted by the activity of the roots show up clearly as white. (After Nye & Tinker, 1977.) autotrophs and heterotrophs EIPC03 10/24/05 1:47 PM Page 78 RESOURCES 79 molecules (proteins, carbohydrates, etc.). These become the resources for heterotrophic organisms (decomposers, parasites, predators and grazers), which take part in a chain of events in which each consumer of a resource becomes, in turn, a resource for another consumer. At each link in this food chain the most obvious distinction is between saprotrophs and predators (defined broadly). Saprotrophs – bacteria, fungi and detritivorous animals (see Chapter 11) – use other organisms, or parts of other organisms, as food but only after they have died, or they consume another organism’s waste or secretory products. Predators use other living organi- sms, or parts of other living organisms, as food. True predators predictably kill their prey. Examples include a moun- tain lion consuming a rabbit but also consumers that we may not refer to as predators in everyday speech: a water flea consuming phytoplankton cells, a squirrel eating an acorn, and even a pitcherplant drowning a mosquito. Grazing can also be regarded as a type of predation, but the food (prey) organism is not killed; only part of the prey is taken, leav- ing the remainder with the potential to regenerate. Grazers feed on (or from) many prey during their lifetime. True predation and grazing are discussed in detail in Chapter 9. Parasitism, too, is a form of predation in which the consumer usually does not kill its food organism; but unlike a grazer, a parasite feeds from only one or a very few host organisms in its lifetime (see Chapter 12). An important distinction amongst animal consumers is whether they are specialized or generalized in their diet. Generalists (polyphagous species) take a wide variety of prey species, though they very often have clear preferences and a rank order of what they will choose when there are alternatives available. Specialists may consume only particular parts of their prey though they range over a number of species. This is most common among herbivores because, as we shall see, different parts of plants are quite different in their composition. Thus, many birds specialize on eating seeds though they are seldom restricted to a particular species. Other speci- alists, however, may feed on only a narrow range of closely related species or even just a single species (when they are said to be monophagous). Examples are caterpillars of the cinnabar moth (which eat the leaves, flower buds and very young stems of species of ragwort, Senecio) and many species of host-specific parasites. Many of the resource-use patterns found among animals reflect the different lifespans of the consumer and what it con- sumes. Individuals of long-lived species are likely to be gener- alists: they cannot depend on one food resource being available throughout their life. Specialization is increasingly likely if a con- sumer has a short lifespan. Evolutionary forces can then shape the timing of the consumer’s food demands to match the timetable of its prey. Specialization also allows the evolution of structures that make it possible to deal very efficiently with par- ticular resources – this is especially the case with mouthparts. A structure like the stylet of an aphid (Figure 3.22) can be interpreted as an exquisite product of the evolutionary process that has given the aphid access to a valuable food resource – or as an example of the ever-deepening rut of specialization that has constrained what aphids can feed on. The more specialized the food resource required by an organism, the more it is constrained to live in patches of that resource or to spend time and energy in searching for it among a mixture of resources. This is one of the costs of specialization. 3.7.1 The nutritional content of plants and animals as food As a ‘package’ of resources, the body of a green plant is quite different from the body of an animal. This has a tremendous effect on the value of these resources as potential food (Figure 3.23). The most important contrast is that plant cells are bounded by walls of cellulose, lignin and/or other structural materials. It is these cell walls that give plant material its high fiber content. The presence of cell walls is also largely responsible for the high fixed carbon content of plant tissues and the high ratio of carbon to other important elements. For example, the carbon : nitrogen (C : N) ratio of plant tissues commonly exceeds 40 : 1, in contrast to the ratios of approximately 10 : 1 in bacteria, fungi and animals. Unlike plants, animal tissues contain no struc- tural carbohydrate or fiber component but are rich in fat and, in particular, protein. The various parts of a plant have very different compositions (Figure 3.23) and so offer quite different resources. Bark, for example, is largely composed of dead cells with corky and lignified walls and is quite useless as a food for most herbivores (even species of ‘bark beetle’ special- ize on the nutritious cambium layer just beneath the bark, rather than on the bark itself ). The richest concentrations of plant pro- teins (and hence of nitrogen) are in the meristems in the buds at shoot apices and in leaf axils. Not surprisingly, these are usually heavily protected with bud scales and defended from herbivores by thorns and spines. Seeds are usually dried, packaged reserves rich in starch or oils as well as specialized storage proteins. And the very sugary and fleshy fruits are resources provided by the plant as ‘payment’ to the animals that disperse the seeds. Very little of the plants’ nitrogen is ‘spent’ on these rewards. The dietary value of different tissues and organs is so differ- ent that it is no surprise to find that most small herbivores are specialists – not only on particular species or plant groups, but on particular plant parts: meristems, leaves, roots, stems, etc. The smaller the herbivore, the finer is the scale of heterogeneity of •••• saprotrophs, predators, grazers and parasites specialists and generalists C : N ratios in animals and plants different plant parts represent very different resources . . . EIPC03 10/24/05 1:47 PM Page 79 82 CHAPTER 3 biomass is increased. If the microbes themselves are carbon- limited, then increased CO2 concentrations, apart from their direct effects on plants, might stimulate microbial activity, making other nutrients, especially nitrogen, available to plants, further stimulating plant growth. Certainly, short-term experiments have demonstrated this kind of effect on decomposer communi- ties. On the other hand, though, decomposers may be nitrogen- limited, either initially or following a period of enhanced plant growth during which nitrogen accumulates in plant biomass and litter. Then, microbial activity would be depressed, diminishing the release of nutrients to plants and potentially preventing their enhanced growth in spite of elevated CO2 concentrations. These, though, are longer term effects and to date very few data have been collected to detect them. The more general issue of local and global ‘carbon budgets’ is taken up again in Section 18.4.6. 3.7.2 Digestion and assimilation of plant material The large amounts of fixed carbon in plant materials mean that they are potentially rich sources of energy. It is other components of the diet (e.g. nitrogen) that are more likely to be limiting. Yet most of that energy is only directly available to consumers if they have enzymes capable of mobilizing cellu- lose and lignins, whereas the overwhelming majority of species in both the plant and animal kingdoms lack these enzymes. Of all the many constraints that put limits on what living organisms can do, the failure of so many to have evolved cellulolytic enzymes is a particular evolutionary puzzle. It may be that gut-inhabiting, cellulolytic prokaryotes have so readily formed intimate, ‘symbiotic’ relationships with herbivores (see Chap- ter 13) that there has been little selection pressure to evolve cellulases of their own (Martin, 1991). It is now recognized that a number of insects do indeed produce their own cellulases but the vast majority nevertheless depend on symbionts. Because most animals lack cellulases, the cell wall material of plants hinders the access of digestive enzymes to the contents of plant cells. The acts of chewing by the grazing mammal, cook- ing by humans and grinding in the gizzard of birds allow diges- tive enzymes to reach cell contents more easily. The carnivore, by contrast, can more safely gulp its food. When plant parts are decomposed, material with a high car- bon content is converted to microbial bodies with a relatively low carbon content – the limitations on microbial growth and multi- plication are resources other than carbon. Thus, when microbes multiply on a decaying plant part, they withdraw nitrogen and other mineral resources from their surroundings and build them into their own microbial bodies. For this reason, and because micro- bial tissue is more readily digested and assimilated, plant detritus that has been richly colonized by microorganisms is generally preferred by detritivorous animals. In herbivorous vertebrates the rate of energy gain from different dietary resources is determined by the structure of the gut – in particular, the balance between a well-stirred anterior chamber in which microbial fermentation occurs (AF), a connecting tube in which there is digestion but no fermentation (D), and a posterior fermentation chamber, the colon and cecum (PF). Models of such three-part digestive systems (Alexander, 1991) suggest that large AF, small D and small PF (e.g. the ruminant) would give near-optimal gains •••• Nutrient mineralization Microbial grazing Plant N uptake Plant growth Microbial growth Soil C availability Plant growth Litter decomposition Nutrient mineralization Plant growth Microbial activities Microbial activities Microbial growth Soil N availability C storage N becomes limiting to microbes Net N accumulation in plant biomass or litter C is limiting to microbes Elevated [CO2] Positive feedback to plant growth Negative feedback to plant growth Time Figure 3.24 Potential positive and negative feedback from elevated CO2 concentrations to plant growth, to microbial activity and back to plant growth. The arrows between descriptors indicate causation; the black arrows alongside descriptors indicate increases or decreases in activity. The dashed arrow from elevated [CO2] to plant growth indicates that any effect may be absent as a result of nutrient-limitation. (After Hu et al., 1999.) cellulases, which most animals lack the gut structures of herbivorous vertebrates EIPC03 10/24/05 1:47 PM Page 82 RESOURCES 83 from poor-quality food, and that large PF, as in horses, is more appropriate for food with less cell wall material and more cell con- tents. For very high-quality food (a very high proportion of cell contents and little cell wall material) the optimum gut has long D and no AF or PF. Elephants, lagomorphs and some rodents eat their own feces and so double the distance traveled by the food resource through the digestive system. This allows further fermentation and digestion but may also allow time for dietary deficiencies (e.g. of vitamins) to be made good by microbial synthesis. These issues are picked up again in Section 13.5. 3.7.3 Physical defenses All organisms are potentially food resources for others and so it is not surprising that many organisms have evolved physical, chemical, morphological and/or behavioral defenses that reduce the chance of an encounter with a consumer and/or increase the chance of surviving such an encounter. But the interaction does not necessarily stop there. A better defended food resource itself exerts a selection pressure on consumers to overcome that defense; though in overcoming that defense, rather than the defenses of other species, the consumer is likely to become relatively specialized on that resource – which is then under particular pressure to defend itself against that particular consumer, and so on. A continuing interaction can therefore be envisaged in which the evolution of both the consumer and the organism consumed depend crucially on the evolution of the other: a coevolutionary ‘arms race’ (Ehrlich & Raven, 1964), which, in its most extreme form, has a coadapted pair of species locked together in perpetual struggle. Of course, the resources of green plants (and of autotrophs in general) are not alive and cannot therefore evolve defenses. Coevolution is also not possible between decomposer organisms and their dead food resources, although bacteria, fungi and detri- tivorous animals will often have to contend with the residual effects of physical and, in particular, chemical defenses in their food. Simple spines can be an effective deterrent. The spiny leaves of holly are not eaten by oak eggar moth larvae (Lasiocampa quercus), but if the spines are removed the leaves are eaten readily. No doubt a similar result would be achieved with foxes as predators and de-spined hedgehogs as prey. In many small planktonic invertebrates that live in lakes, the development of spines, crests and other appendages that reduce their vulnerabil- ity to predation can be induced by a predator’s presence. Thus, for example, spine development in the offspring of brachionid rotifers, including Keratella cochlearis, is promoted if their mother was cultured in a medium conditioned by the predatory rotifer, Asplachna priodonta (Stemberger & Gilbert, 1984; Snell, 1998). At a smaller scale still, many plant surfaces are clothed in epidermal hairs (trichomes) and in some species these develop thick secondary walls to form strong hooks or points that may trap or impale insects. Any feature that increases the energy a consumer spends in discover- ing or handling a food item – the thick shell of a nut or the fibrous cone on a pine – is a defense if, as a consequence, the consumer eats less of it. The green plant uses none of its energetic resources in running away and so may have relatively more available to invest in energy-rich defense struc- tures. Moreover, most green plants are probably relatively over- provided with energy resources, making it cheap to build shells around seeds and woody spines on stems – mainly out of cellu- lose and lignin – and so protecting the real riches: the scarce resources of nitrogen, phosphorus, potassium, etc. in the embryos and meristems. Seeds are most at risk to predators when they have just ripened and are still attached, in a cone or ovary, to the parent plant, but their value is literally dissipated as soon as the capsule opens and the seeds are shed. The poppies illustrate this point. The seeds of wild poppies are shed through a series of pores at the apex of the capsule as it waves in the wind. Two of the species, Papaver rhoeas and P. dubium, open these pores as soon as the seed is ripe and the capsules are often empty by the following day. Two other species, P. argemone and P. hybridum, have seeds that are large relative to the size of the capsule pores and dispersal is a slow process over the fall and winter months. The capsules of these species are defended by spines. The cultiv- ated poppy (P. somniferum) by contrast, has been selected by humans not to disperse its seeds – the capsule pores do not open. Birds can therefore be a serious pest of the cultivated poppy; they tear open the capsules to reach an oil- and protein-rich reward. Humans, in fact, have selected most of their crops to retain rather than disperse their seeds and these represent sitting targets for seed-eating birds. 3.7.4 Chemical defenses The plant kingdom is very rich in chemicals that apparently play no role in the normal pathways of plant bio- chemistry. These ‘secondary’ chemicals range from simple molecules like oxalic acid and cyanide to the more complex glu- cosinolates, alkaloids, terpenoids, saponins, flavonoids and tannins (Futuyma, 1983). Many of these have been shown to be toxic to a wide range of potential consumers. For example, populations of white clover, Trifolium repens, are commonly polymorphic for the ability to release hydrogen cyanide when the tissues are attacked. Plants that lack the ability to generate hydrogen cyanide are eaten by slugs and snails: the cyanogenic forms are nibbled but then rejected. Many researchers have assumed that •••• coevolution spines shells seeds: dissipation or protection secondary chemicals: protectants? EIPC03 10/24/05 1:47 PM Page 83 84 CHAPTER 3 protection against consumers has provided the selective pressure favoring the production of such chemicals. Many others, however, have questioned whether the selective force of herbivory is powerful enough for this (their production may be costly to the plants in terms of essential nutrients) and have pointed to other properties that they possess: for example as protectants against ultraviolet radiation (Shirley, 1996). None the less, in the few cases where selection experiments have been carried out, plants reared in the presence of consumers have evolved enhanced defenses against these enemies, relative to control plants reared in the absence of consumers (Rausher, 2001). Later, in Chapter 9 when we look in more detail at the interaction between predators and their prey, we will look at the costs and benefits of prey (especially plant) defense to both the prey itself and its con- sumers. Here, we focus more on the nature of those defenses. If the attentions of herbivores select for plant defensive chemicals, then equally, those chemicals will select for adaptations in herbivores that can overcome them: a classic coevolutionary ‘arms race’. This, though, suggests that plants should become ever more noxious and herbivores ever more specialized, leaving unanswered the question of why there are so many gen- eralist herbivores, capable of feeding from many plants (Cornell & Hawkins, 2003). An answer has been suggested by ‘apparency theory’ (Feeny, 1976; Rhoades & Cates, 1976). This is based on the observation that noxious plant chemicals can be classified broadly into two types: (i) toxic (or qualitative) chemicals, which are poisonous even in small quantities; and (ii) digestion-reducing (or quantitative) chemicals, which act in proportion to their con- centration. Tannins are an example of the second type. They bind proteins, rendering tissues such as mature oak leaves relatively indigestible. The theory further supposes that toxic chemicals, by virtue of their specificity, are likely to be the foundation of an arms race, requiring an equally simple and specific response from a herbivore; whereas chemicals that make plants generally indigestible are much more difficult to overcome. Apparency theory then proposes that relatively short-lived, ephemeral plants (said to be ‘unapparent’) gain a measure of pro- tection from consumers because of the unpredictability of their appearance in space and time. They therefore need to invest less in defense than predictable, long-lived (‘apparent’) species like forest trees. Moreover, the apparent species, precisely because they are apparent for long, predictable periods to a large number of herbivores, should invest in digestion-reducing chemicals that, while costly, will afford them broad protection; whereas unapparent plants should produce toxins since it is only likely to pay a few special- ist species to coevolve against them. Apparency theory, incorporating ideas on coevolution, there- fore makes a number of predictions (Cornell & Hawkins, 2003). The most obvious is that more unapparent plants are more likely to be protected by simple, toxic compounds than by more com- plex, digestion-inhibiting compounds. This can even be seen in the changing balance of chemical defense in some plants as the season progresses. For example, in the bracken fern (Pteridium aquil- inum), the young leaves that push up through the soil in spring are less apparent to potential herbivores than the luxuriant foliage in late summer. The young leaves are rich in cyanogenic glucosinolates, whilst the tannin content steadily increases in concentration to its maximum in mature leaves (Rhoades & Cates, 1976). A more subtle prediction of the theory is that specialist herbivores, having invested evolutionarily in overcoming par- ticular chemicals, should perform best when faced with those chemicals (compared to chemicals they would not normally encounter); whereas generalists, having invested in performing well when faced with a wide range of chemicals, should per- form least well when faced with chemicals that have provoked coevolutionary responses from specialists. This is supported by an analysis of a wide range of data sets for insect herbivores fed on artificial diets with added chemicals (892 insect/chemical combinations) shown in Figure 3.25. Furthermore, plants are predicted to differ in their chemical defenses not only from species to species but also within an individual plant. ‘Optimal defense theory’ predicts that the more important an organ or tissue is for an organism’s fitness, the better protected it will be; and in the present context, it predicts that more important plant parts should be protected by constitu- tive chemicals (produced all the time), whereas less important parts should rely on inducible chemicals, only produced in response to damage itself, and hence with far lower fixed costs to the plants (McKey, 1979; Strauss et al., 2004). This is confirmed, for example, by a study of wild radish, Raphanus sativus, in which plants were either subjected to herbivory by caterpillars of the butterfly, Pieris rapae, or left as unmanipulated controls (Strauss et al., 2004). Petals (and all parts of the flower) are known in this insect- pollinated plant to be highly important to fitness. Concentrations of protective glucosinolates were twice as high in petals as in undamaged leaves, and these levels were maintained constitutively, irrespective of whether the petals were damaged by the caterpil- lars (Figure 3.26). Leaves, on the other hand, have a much less direct influence on fitness: high levels of leaf damage can be sustained without any measurable effect on reproductive output. Constitutive levels of glucosinolates, as already noted, were low; but if the leaves were damaged the (induced) concentrations were even higher than in the petals. Similar results were found for the brown seaweed, Sargassum filipendula, where the holdfast at its base was the most valuable tissue: without it the plant would be cast adrift in the water (Taylor et al., 2002). This was protected by costly constitutive, quantitat- ive chemicals, whereas the much less valuable youngest stipes (effectively stems) near the tip of the plant were protected only by toxic chemicals induced by grazing. •••• apparency theory optimal defense theory: constitutive and inducible defenses EIPC03 10/24/05 1:47 PM Page 84